new management of capital flowsunctad.org/en/docs/gdsmdpb20031p3_en.pdf · 2012. 6. 18. ·...

UNCTAD/GDS/MDPB/2003/1

UNITED NATIONS CONFERENCE ON TRADE AND DEVELOPMENT Geneva

Management of Capital Flows: Comparative experiences and implications for Africa

PART 3.

UNITED NATIONS New York and Geneva, April 2003

MANAGEMENT OF THE CAPITAL ACCOUNT: A STUDY OF INDIA AND MALAYSIA 109

MANAGEMENT OF THE CAPITAL ACCOUNT:A STUDY OF INDIA AND MALAYSIA

Indira Rajaraman

I. INTRODUCTION

A. India and Malaysia

This paper contrasts the approaches of India and Malaysia to capitalaccount management, which, although widely different historically, havearrived at a remarkable convergence after the East Asian crisis of 1997.The paper does not compare Malaysia with its East Asian neighbours interms of post-1997 outcomes or in terms of their relative situations at thetime of the crisis, as this ground is already covered in a large body ofliterature (see for example, Borgini, Claessens and Ferri, 2000; Das, 1999;Furman and Stiglitz, 1998; Radelet and Sachs, 1998).

India and Malaysia are macroeconomically quite disparate. Table 1presents a summary macroeconomic profile that highlights the vastdifferences between the two. Beyond the more obvious differences ineconomic size in terms of total and per capita GDP, there is the historicalgrowth differential. India took birth in 1991 as an open economy after30 years of slow autarkic growth at 3.5 per cent, and some speeding up to

INDIRA RAJARAMAN110Ta

ble

1

MA

CR

OEC

ON

OM

IC P

RO

FILE

FO

R M

ALA

YSIA

AN

D IN

DIA

, 199

7–19

99

Mal

aysi

aIn

dia

1997

1998

1999

1997

1998

1999

GN

Pm

p ($ b

illio

n)94

.868

.673

.541

7.2

426.

544

1.8

Popu

latio

n (m

illio

n)21

.722

.222

.796

2.4

979.

799

7.0

Per c

apita

GN

P (c

urre

nt $

)4

369

3 09

03

237

434

435

443

PPP

(cur

rent

$)

8 55

57

699

7 64

02

019

2 06

02

230

Fore

ign

rese

rves

($ b

illio

n)15

.626

.230

.929

.432

.538

.0Im

port

cove

r (m

onth

s)3.

45.

75.

96.

98.

28.

2

Gro

wth

rate

of G

DP

(rea

l)7.

3-7

.45.

65.

06.

86.

4

Shar

e of

GD

P (p

er c

ent)

Gro

ss d

omes

tic in

vest

men

t43

.026

.622

.326

.223

.423

.3G

ross

dom

estic

sav

ing

43.9

48.5

47.3

24.7

22.3

22.3

Expo

rts (g

oods

& n

on-fa

ctor

ser

vice

s)93

.311

5.2

121.

711

.211

.512

.1

Exte

rnal

deb

t /G

DP

(per

cen

t)60

.656

.757

.824

.423

.522

.0Sh

ort-t

erm

/tota

l ext

erna

l25

.319

.915

.05.

44.

54.

1D

ebt-s

ervi

ce ra

tio5.

56.

75.

819

.018

.016

.0

Fisc

al (p

er c

ent)

Cur

rent

bal

ance

/GD

P7.

54.

34.

0-3

.1-3

.9-3

.8

Infla

tion

rate

(CPI

) (pe

r cen

t)2.

75.

32.

87.

213

.25.

0

Sour

ce:

Dat

a on

GN

P in

cur

rent

and

PPP

$ fr

om W

orld

Ban

k, W

orld

Dev

elop

men

t Ind

icat

ors

2000

; all

othe

r dat

a fro

m c

ount

ry s

ourc

es. F

or M

alay

sia

from

Ban

k N

egar

aM

alay

sia

(BN

M),

Annu

al R

epor

t 199

9, a

nd M

onth

ly S

tatis

tical

Bul

letin

(Jun

e 20

00);

for I

ndia

from

Res

erve

Ban

k of

Indi

a (R

BI),

Ann

ual R

epor

t 200

0, a

nd B

ulle

tin(A

ugus

t 200

0) (e

xcep

t inf

latio

n; s

ee n

otes

to ta

ble

2).

Not

e:In

dian

figu

res

perta

in to

the

Indi

an fi

scal

yea

r, w

hich

runs

from

1 A

pril

to 3

1 M

arch

. The

deb

t sto

ck fi

gure

s (s

hort-

term

sha

re) c

orre

spon

ding

ly p

erta

in to

the

end

of th

e fis

cal y

ear (

i.e. e

nd-M

arch

200

0) fo

r the

yea

r 199

9. T

he fi

scal

bal

ance

for I

ndia

cov

ers

the

cent

ral G

over

nmen

t alo

ne. C

PI=

cons

umer

pric

e in

dex.


an average growth rate of 5.6 per cent in the period 1980–1990, the decadejust prior to the big-ticket reforms of the 1990s. By contrast in Malaysia, amember of the second-rank quadruplet of Asian tigers, no decade prior tothe1990s saw average growth under 5.2 per cent and no year in the 1990shad growth below 8.9 per cent until 1997, when the rate fell to 7.3 percent. Supporting this historical growth disparity are vast differences insavings as a per cent of GDP, at the 25 per cent mark in India, and at nearlythe 50 per cent mark in Malaysia. The investment rate, historically high inMalaysia but below the savings rate (the obverse of a historically positivebalance on trade in goods and non-factor services) has come down in thepost-crisis years to 22 per cent in 1999, with a corresponding widening ofthe favourable balance of trade. In India, the investment rate lies typicallyat about 1.5 percentage points above the savings rate, the obverse of ahistorically negative balance of trade. The Malaysian Government contri-buted to domestic savings with positive current fiscal balances, whichcontinued throughout the crisis and post-crisis years; in India on the otherhand, the Government dissaves, with negative current fiscal balances onthe central Government account alone amounting to between 3 and 4 percent of GDP.

Clearly the most startling difference between the two economies liesin their degree of openness. Exports in Malaysia have traditionally beenroughly equivalent to GDP, and have actually exceeded GDP in the post-crisis years; India on the other hand had an export/GDP ratio at the11–12 per cent level during the period 1997–1999, a notch up from ratiosusually well below 10 per cent prior to that period.

The Malaysian tradition of openness goes back a long way. Malaysiaaccepted Article VIII status under the IMF (current account convertibility)as far back as November 1968. The Malaysian ringgit was floated at thesame time as the major world currencies, in 1973 (see appendices A and B,which list external policy milestones in India and Malaysia respectively).India, on the other hand, accepted Article VIII status only in August 1994,following a unified floating rate for the rupee in March 1993.

Progress towards capital account convertibility in Malaysia began withthe ringgit float in 1973, with a second major thrust in 1987. By 1990,

INDIRA RAJARAMAN112

there was even an offshore over-the-counter (OTC) market for Malaysianshares in Singapore which reinforced a pre-existing offshore ringgit market,perhaps the only case of an offshore market for a developing countrycurrency. In India, by contrast, capital account convertibility has neverexisted for residents, although non-resident capital inflows are fullyconvertible. The 1997 Reserve Bank of India (Tarapore) Committee recom-mendation of a phased move to full capital account convertibility by 2000(Reserve Bank of India (RBI), 1997) was not adhered to on account offailure to meet its recommended prudential markers on the fiscal andinflation fronts.

The Malaysian response to the currency crisis of 1997 backtrackedon what had been perhaps the longest tradition of openness on current andcapital accounts in the developing world. As the capital outflow crisis inEast Asia hit Malaysia with full force (although Malaysia was inexplicablyexcluded by Summers (2000) from the list of Asian countries affected bythe crises), the country famously imposed capital exit barriers and a policyof lower interest rates that ran directly orthogonal to the policy in otherEast Asian countries similarly afflicted.

By contrast, the Indian external liquidity crisis in mid-1991, whenforeign reserves fell to an all-time low of $1 billion, led to the opening upof the Indian capital account. Following the opening up of the Indianeconomy in 1991, there was a period, mid-decade, when the principalproblem of capital account management arose from too large an inflowrather than too little. This was a problem in Malaysia as well at around thesame period, handled through temporary entry barriers that were soondismantled as the situation turned.

The East Asian crisis of 1997 reaffirmed the possibility of multipleequilibria in currency markets. The issue of whether hedge funds played arole in the crisis is empirically unproven, and is less important than thestrenuous attempts made to investigate the matter might initially suggest.The finding by Brown, Goetzmann and Park (1998), of an absence of anycorrelation between hedge fund short positions in Asian currencies andmovements in exchange rates, had methodological flaws, which yieldedimplausible overestimates of hedge fund short positions. What these


attempts have served to highlight is the severe paucity of information onhedge funds. Reported figures suggest that hedge funds may be less gearedthan commercial and investment banks, but it is well-known that hedgefunds rely principally on off-balance-sheet techniques to obtain leverage.

Even if market manipulation in its many forms – whether action-based(insider trading), information-based (rumour spreading), or trade-based(cornering markets) – is ruled out as a contributory factor, it is clear thatherd behaviour with positive feedback can precipitate a crash, with sellinglow inducing further herd selling and therefore becoming profitable. Thiscan then be transmitted across markets through pure contagion, where goodmacro-policy and capital markets, characterized by high secondaryliquidity, are no guarantee against infection. On the contrary, high liquiditymay accentuate amenability to infection, as indeed was the case in Malaysia.

B. The issues

This study examines what has been done in both India and Malaysiain terms of controlling the quantum and the composition of capital flows.There are two main relevant issues in composition: debt/equity composition,and the maturity structure of external debt.

In terms of a hierarchy of desired forms of capital inflows, equityinvestment is thought superior to debt inflows because of the sharing ofrisk (for countries as for corporates); and within equity flows, directinvestment is superior to portfolio investment because of the greater stakein the host country. But between portfolio and debt, the relative hierarchyis less clear. The short-term component of external debt is by no means acomplete measure of the scope for short-term capital outflows. Defendantsof portfolio capital tend to dismiss its potential volatility as low, on thegrounds that downward revisions in expectations regarding stocks impingeequally on domestic and foreign investors, so that changes in expectationswould be fully reflected in downward stock market valuations, and not atall (or very little) in foreign investor sale and exit. The East Asian crisishas dramatically shown the falsity of that position. Such a position in one

INDIRA RAJARAMAN114

market leads to pressure on foreign investors to liquidate in third countries,even when stock expectations remain high, resulting in outward capitalflows from these third countries by contagion.

When potential outward capital flows by residents are added on topotential foreign investor outflows, it is clear that the proportion of short-term debt in total external debt is not so much a precipitator of capitaloutflows, as an indicator of how sharp the spike can be, whether precipitatedby contagion or real factors.

In 1999, external debt as a per cent of GDP was higher in Malaysia, at58 per cent (down from 61 per cent at the end of 1997), than in India at22 per cent (table 1). Also, although the Malaysian short-term debt as apercentage of total external debt declined from 25 per cent in 1997 to 15 percent in 1999, it was nevertheless much higher than the Indian short-termdebt of 4 per cent. However, external debt taken as a percentage of exportsalters somewhat the overall picture of greater conservatism in India thanin Malaysia in respect of external borrowing (table 2). India until 1999was above the prudent thumbrule of 200 per cent. Malaysia on the otherhand had an external debt/exports ratio of 44 per cent. The figures forshort-term external debt as a percentage of exports are even more startling:Malaysia’s at 5 per cent was far lower than India’s at 37 per cent at the startof the decade, and, despite a rise thereafter, exceeded India’s only in 1997.

Thus, normalized by exported tradeables rather than by the size ofthe economy aggregating across tradables and non-tradables, Malaysiadisplays greater external prudence than India over the 1990s. The debtservice ratio in Malaysia is correspondingly far lower, despite a much lowerconcessional proportion in total external debt in Malaysia (6 per cent) ascompared to India (41 per cent), a result of the far greater proportion ofexports in GDP (table 1).

Table 2 shows that in the 1990s Malaysia sustained stable annual flowsof FDI in the $2–4 billion range, whereas India saw a very slow climb tolevels of about $2 billion annually. Both countries saw a sharp increasein portfolio inflows mid-decade, which remained subsequently at the$3–4 billion level in India, but turned into net outflows of $10 billion in

MANAGEMENT OF THE CAPITAL ACCOUNT: A STUDY OF INDIA AND MALAYSIA 115Ta

ble

2

EXTE

RN

AL

IND

ICAT

OR

S FO

R M

ALA

YSIA

AN

D IN

DIA

IN T

HE

1990

s

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

Fore

ign

dire

ct in

vest

men

t ($

billi

on, g

ross

)M

alay

sia

....

..1.

83.

53.

83.

72.

41.

91.

8In

dia

0.2

0.1

0.3

0.6

1.0

2.1

2.4

3.6

2.6

2.2

Fore

ign

portf

olio

inve

stm

ent (

$ bi

llion

, net

)M

alay

sia

-1.0

-0.7

3.0

9.9

5.4

2.1

4.4

-10.

1-0

.30.

3In

dia

0.3

1.4

0.0

2.3

4.7

1.8

4.2

4.0

4.5

3.0

Exte

rnal

deb

t/exp

orts

of g

oods

and

non-

fact

or s

ervi

ces

(per

cen

t)M

alay

sia

51.8

48.5

48.9

50.9

42.3

40.6

42.1

65.0

49.2

44.0

Indi

a36

3.5

366.

838

2.7

337.

631

0.7

238.

022

4.7

209.

120

7.2

181.

8

Shor

t-ter

m d

ebt/e

xpor

ts o

f goo

ds a

ndno

n-fa

ctor

ser

vice

s (p

er c

ent)

Mal

aysi

a5.

06.

811

.512

.88.

27.

710

.816

.59.

86.

7In

dia

37.1

30.4

26.9

13.0

12.9

12.7

16.2

11.2

9.1

7.5

Exch

ange

rate

(per

iod

aver

age

LCU

/$)

Mal

aysi

a2.

72.

82.

62.

62.

62.

52.

52.

83.

93.

8In

dia

17.5

22.7

25.9

30.5

31.4

32.4

35.4

36.3

41.3

43.0

Sour

ce:

Wor

ld B

ank,

Wor

ld D

evel

opm

ent

Indi

cato

rs,

2000

, fo

r In

dia

alon

e, f

or p

roce

ssed

cal

enda

r ye

ar e

stim

ates

fro

m I

ndia

n fis

cal y

ear

(Apr

il-M

arch

) da

ta,

com

para

ble

to th

ose

for M

alay

sia,

whi

ch re

ports

by

cale

ndar

yea

r.N

ote:

Mal

aysi

an c

ount

ry s

ourc

es (s

ee s

ourc

es fo

r tab

le 1

) con

flict

ed s

erio

usly

with

the

Wor

ld B

ank

data

and

wer

e th

eref

ore

chos

en in

all

case

s. D

ata

on F

DI i

n th

efir

st th

ree

year

s w

ere

not a

vaila

ble.

LCU

= lo

cal c

urre

ncy

unit.

INDIRA RAJARAMAN116

1997 in Malaysia. Given the footloose character of portfolio flows, andgiven also the negative real impact of an interest rate hike, which is thestandard macroeconomic market corrective for capital outflows, it is notsurprising that exit barriers to capital outflows have become, and willremain, a component of the menu of policy options to stabilize currencymarkets. The case of Malaysia has incontrovertibly established the realeconomy benefits of temporary suspensions of capital convertibility (seealso Athukorala,1998; and James, 1999).

C. Outline of the paper

In what follows, the approach to capital account management in Indiais examined in section II, and in Malaysia in section III. Section IV presentsconclusions. Appendices A and B list major external policy milestones inIndia and Malaysia respectively.

II. INDIA

A. Introduction

The opening of the Indian economy began in July 1991, in immediateresponse to an external liquidity crisis, with foreign reserves falling to alow of $1 billion and a longer-term dissatisfaction with 40 years of slow,autarkic, growth. Prior to reform, the Indian rupee was not convertible oncurrent account, and was officially set at an undisclosed moving peg to abasket of currencies (see Rajaraman, 1991). Import tariff rates were amongthe most punitive in the world, and capital inflows were largely officialborrowings on concessional terms. Even today concessional debt accountsfor 40 per cent of total external debt. Foreign portfolio investment was notpermissible, and even FDI was regarded with suspicion and permissibleonly through case-by-case approval.


Since 1991, the external price of the rupee has become market-determined, albeit with periods of intensive intervention by the ReserveBank of India (RBI). There is current account convertibility, with caps oninvisible outflows that vary by category, and there has been increasingfacilitation of foreign direct and portfolio investment. Capital outflowsfrom residents are still prohibited, although some liberalization of outwardFDI by corporates is now possible without prior approval, through shareswaps up to specified limits, and in excess of these limits with specialapproval. All outflows of capital from non-corporate residents remainbanned. Foreign capital inflows, on the other hand, are granted full freedomof repatriation, barring a few channels of inflows for non-resident Indiansthat are non-repatriable. Thus there is capital account convertibility fornon-residents but not for residents.

Since 1 June 2000, the legal regime governing both current and capitalaccount transactions is defined by the Foreign Exchange Management Act(FEMA) of 1999, which replaced the Foreign Exchange Regulation Act(FERA) of 1973. FERA restrictions were not on volume of capital inflowsso much as the denial of parity of treatment in terms of freedom offunctioning between “FERA” companies (those with more than 40 per centforeign equity) and other companies, and there were criminal penalties forcontravention of the ban on foreign exchange outflows from residents.

FEMA is a civil law, a major departure from FERA, and it does notapply to Indian citizens resident outside India, unlike FERA, which appliedto all Indian citizens regardless of the location of their residence. Thus,the passage of FEMA marks a major legal milestone in the opening up ofthe Indian economy. FEMA has a few residual restrictions on currentaccount transactions (7 banned categories; 16 requiring permission only ifa ceiling amount is crossed; and 11 needing permission irrespective ofamount). On the capital account, the FERA amendment of 1993 had alreadylegally enabled liberalization in respect of FDI and portfolio inflows thatwas very gradually introduced over the 1990s. FEMA introduced first-time liberalization of rules in respect of cross-border inheritance of property,and first-time automatic approval for outward FDI by corporates throughstock swap options, subject to sectoral caps, and with possible approvalbeyond those caps.

INDIRA RAJARAMAN118

In what follows, section II.B reviews the structure of the balance ofpayments (BOP) as it has evolved over the 1990s. Section II.C examinesin further detail current invisibles, which have kept the current accountdeficit at well under 1.5 per cent of GDP for all but one year in the decade,and reduced the need for capital inflows. Section II.D examines capitalaccount inflows in respect of foreign equity investment and FDI,commercial borrowings, and special deposit schemes for non-residentIndians (NRIs) – a peculiar feature of the Indian capital account. SectionII.E examines the return of flight capital during the 1990s, motivated bythe progressive liberalization of the current account. Section II.F examinesthe market for foreign exchange in India, and the anatomy of interventionsby the RBI in that market. There are no convertibility restrictions on foreigncapital; thus the willingness of foreign capital to enter is a function of thefundamentals of the Indian economy, and of its political and economicgovernance parameters. A key feature of this, the health of the financialsector, is addressed in section II.G. Section II.H concludes.

Appendix A lists Indian external policy milestones in the 1990s. Indianpolicy in any sphere is characterized by incremental dribble, withpunctuation at frequent intervals; the opening up in the 1990s shares thischaracteristic, although fortunately the moves are directionally monotonic.Instead of listing every change, which would render the Appendix tedious,we record only those which altered the essential parameters of policy, soas to make possible an understanding of the intent of the process at a glance.

B. The balance of payments: an overview

The first notable feature about the Indian current account after 1990/91,the crisis year preceding the introduction of reforms in July 1991, is thatthe deficit on current account has been at worst a little over 1.5 per cent ofGDP, usually well under (table 3). This has been not because of a lowdeficit on trade account, which had grown over the period to the region of$17 billion by 1999/00, but because of an explosion of net invisible inflowsstarting 1994/95, to the region of $13 billion by 1999/00. Clearly the


ble

3

IND

IA: O

VER

ALL

BA

LAN

CE

OF

PAYM

ENTS

, 199

0–20

00($

bill

ion)

Item

1990

/91

1991

/92

1992

/93

1993

/94

1994

/95

1995

/96

1996

/97

1997

/98

1998

/99

1999

/00

I.M

erch

. tra

de b

alan

ce (A

-B)

-9.4

4-2

.80

-5.4

5-4

.06

-9.0

5-1

1.36

-14.

82-1

5.51

-13.

25-1

7.10

A.

Exp

orts

, f.o

.b.

18.4

818

.27

18.8

722

.68

26.8

632

.31

34.1

335

.68

34.3

038

.29

B.

Impo

rts, c

.i.f.

27.9

121

.06

24.3

226

.74

35.9

043

.67

48.9

551

.19

47.5

455

.38

II.In

visi

bles

, net

-0.2

41.

621.

922.

905.

685.

4510

.20

10.0

19.

2112

.94

III.

Cur

rent

acc

ount

(I+I

I)-9

.68

-1.1

8-3

.53

-1.1

6-3

.37

-5.9

1-4

.62

-5.5

0-4

.04

-4.1

6/G

DP

(per

cen

t)-(

3.20

)-(

0.40

)-(

1.80

)-(

0.40

)-(

1.00

)-(

1.60

)-(

1.20

)-(

1.30

)-(

1.00

)-(

0.90

)IV

.C

apita

l Acc

ount

(A to

F),

net

7.19

3.78

2.94

9.70

9.16

4.69

11.4

110

.01

8.26

10.5

6A

.Fo

reig

n in

vest

men

t0.

100.

130.

564.

244.

814.

816.

155.

392.

415.

19B

.E

xter

nal a

ssis

tanc

e2.

213.

041.

861.

901.

530.

881.

110.

910.

820.

89C

.C

omm

erci

al b

orro

win

gs (m

& lt

)2.

251.

46-0

.36

0.61

1.03

1.28

2.85

4.00

4.36

0.33

D.

Rup

ee d

ebt s

ervi

ce-1

.19

-1.2

4-0

.88

-1.0

5-0

.98

-0.9

5-0

.73

-0.7

7-0

.80

-0.7

1E

.N

RI d

epos

its1.

540.

292.

001.

210.

171.

103.

351.

131.

742.

14F.

1. S

hort

term

trad

e cr

edits

1.08

-0.5

2-1

.08

-0.7

70.

390.

050.

84-0

.10

-0.7

50.

382.

Oth

er c

apita

l1.

200.

620.

833.

572.

21-2

.48

-2.1

6-0

.54

0.48

2.42

V.O

vera

ll B

alan

ce (I

II+IV

)-2

.49

2.60

-0.5

98.

545.

79-1

.22

6.79

4.51

4.22

6.40

VI.

Res

erve

s an

d re

late

d ite

ms

2.49

-2.6

00.

59-8

.54

-5.7

91.

22-6

.79

-4.5

1-4

.22

-6.4

0

IV(A

+C+E

+F)/I

II (p

er c

ent)

(64)

(168

)(5

5)(7

64)

(256

)(8

1)(2

39)

(179

)(2

04)

(251

)

Sour

ce:

Res

erve

Ban

k of

Indi

a, H

andb

ook

of S

tatis

tics

on In

dian

Eco

nom

y, 1

999,

tabl

es 1

22 a

nd 1

95 a

nd R

BI B

ulle

tin, a

ssor

ted

issu

es.

Not

e:S

tarti

ng in

199

0/91

, de

fenc

e im

ports

are

rec

orde

d in

the

yea

r of

impo

rt, w

ith m

ediu

m a

nd lo

ng t

erm

fin

anci

ng c

redi

ts in

clud

ed in

ext

erna

l com

mer

cial

borr

owin

gs; w

hen

defe

nce

impo

rts a

re g

row

ing

over

tim

e, th

e im

port

bill

in th

e im

porti

ng y

ear i

s la

rger

than

whe

n re

cord

ed a

t the

tim

e of

ser

vici

ng th

e cr

edit.

Oth

er c

hang

es m

ade

afte

r 199

0/91

do

not i

mpa

ct u

pon

the

capi

tal a

ccou

nt (s

ee n

otes

to ta

ble

5). “

Rup

ee d

ebt s

ervi

ce” (

item

IV.D

) is

in re

spec

t of r

upee

-de

nom

inat

ed e

xter

nal a

ssis

tanc

e fro

m th

e fo

rmer

CO

ME

CO

N, a

nd in

clud

es in

tere

st. S

hort-

term

trad

e cr

edits

exc

lude

loca

l cur

renc

y-ba

cked

cre

dit t

hrou

ghba

nkin

g ch

anne

ls (s

ee te

xt).

Oth

er c

apita

l inc

lude

s er

rors

and

om

issi

ons.

INDIRA RAJARAMAN120

composition of invisibles is a matter of some importance for an assessmentof whether this growth is sustainable (section II.C).

The low deficit on current account has meant that in most years of theperiod non-official capital flows have well exceeded 100 per cent of theaccount deficit (table 3). Thus, even though net official aid – bothconvertible and rupee-denominated components – has been negligible,except in the early 1990s, the decade has seen a rise in foreign reserves, toa stock of nearly $30 billion at the end of 1998/99 (table 4). This has beenthe single most important indicator of the success of the economic reformprogramme begun in July 1991. Indeed, for a period during the mid-1990s,the chief problem of capital account management arose from too large aninflow rather than too little.

Table 4

INDIA: EXTERNAL RESERVES, 1990–2000($ billion)

Foreign currency assets

End-March Total SDRs Gold Total Change

1990/91 5.83 0.10 3.50 2.24 -1.13

1991/92 9.22 0.09 3.50 5.63 3.40

1992/93 9.83 0.02 3.38 6.43 0.80

1993/94 19.25 0.11 4.08 15.07 8.63

1994/95 25.19 0.01 4.37 20.81 5.74

1995/96 21.69 0.08 4.56 17.04 -3.77

1996/97 26.42 0.00 4.05 22.37 5.32

1997/98 29.37 0.00 3.39 25.98 3.61

1998/99 32.49 0.01 2.96 29.52 3.55

1999/00 38.03 0.00 2.97 35.06 5.54

Source: Ibid., table 136, supplemented by RBI Bulletin, August 2000.Note: The figures in this table are not consistent with the figures from the same source for balance of

payments flow, because of changes in parities between constituent currencies in foreign assets.


C. Invisibles in the current account

The major prop of the balance of payments in the 1990s has been thespurt of private transfers in the current account to annual inflows of around$12 billion, with two clearly visible discontinuous points of increase.

The first in 1993/94 was on account of the liberalization of goldimports, previously banned. The differential between the international andIndian prices of gold fuelled the infamous hawala channel of conversionof foreign currencies into and out of rupees. Hawala was an unofficial, butefficiently functioning, market whereby rupee assets were convertible intoforeign currencies obtained principally from diverted remittances, but alsofrom misinvoiced trade. With the 1992/93 decision to allow up to 5 kg ofgold per non-resident Indian (NRI) returning to the country, gold importsentered into formally reported imports, with a contra entry in what arenow recorded (implicit) private transfers, thus accounting for the increasein the latter starting in 1993/94. The net direct effect of each such goldimport on the current balance was thus neutral, but secondary effectsimproved the current account balance with the narrowing gap betweenofficial and hawala rates on the dollar, and the consequently reducedattraction of transmitting private transfers through the hawala route. Withthe permissibility, starting October 1997, of gold imports by designatedimporting agents, the domestic price of gold declined still more to paritywith international prices, thus eliminating altogether the market for divertedprivate transfers to finance gold smuggling.

The second discontinuous increase in transfers, starting in 1996/97,is the analogue of a policy change in respect of NRI deposits (see sectionII.E below).1 With the start in 1992/93 of non-repatriable external depositsdenominated in rupees, the first repayments of (non-repatriable) principalthree years down the line in 1996/97 were contra-entered as privatetransfers, although other accounting arrangements might have been devisedfor what are not current account flows. But given this recording convention,private transfers are now in the $10–12 billion range (see table 5). Otherinvisibles sum to zero because of steady outflows of the order of $3 billionon account of interest on external debt. Software exports have grown at

INDIRA RAJARAMAN122

explosive rates from an initial small base to reach present annual flows ofthe order of $4 billion per year.

D. Capital inflows

(i) Equity

The liberalization of FDI, effective as of 24 July 1991, followed soonthereafter by the signing of the Multilateral Investment Guarantee Agency(MIGA) protocol, could be said to mark the start of the big-ticket reformsof the Indian economy. First-time automatic approval up to 51 per centequity in 34 sectors was a major departure from the earlier regime of case-by-case consideration, with equity shares rarely permitted beyond a ceiling

Table 5

INDIA: INVISIBLES BY CATEGORIES OF TRANSACTIONS, 1990–2000($ billion)

Non-factor Factor Private Official Invisibles,services, net services, net transfers, net transfers, net net

Year (I) (II) (III) (IV) (I to IV)

1990/91 0.98 -3.75 2.07 0.46 -0.241991/92 1.21 -3.83 3.78 0.46 1.621992/93 1.13 -3.42 3.85 0.36 1.921993/94 0.54 -3.27 5.27 0.37 2.901994/95 0.60 -3.43 8.09 0.42 5.681995/96 -0.20 -3.21 8.51 0.35 5.451996/97 0.73 -3.31 12.37 0.41 10.201997/98 1.32 -3.52 11.83 0.38 10.011998/99 2.17 -3.54 10.28 0.31 9.211999/00 3.86 -3.56 12.26 0.38 12.94

Source: Ibid., table 124, supplemented by RBI Bulletin, August 2000.Note: Starting 1992/93, data on gold and silver brought in by Indian returning from abroad have been

included under import payments with a corresponding entry in private transfers. Starting 1996/97,repayments of non-resident non-repatriable rupee accounts were contra-entered as privatetransfers in the current account (see RBI Annual Report 1998-99, Box VI.3). Both changes reflectpolicy changes, and therefore do not introduce discontinuities in the data series.


of 40 per cent, and, when permitted, subject to draconian limits on freedomof functioning prescribed under the FERA of 1973. These limits were doneaway with in January 1993 through amendment of the FERA.

Regulation of capital markets without government control was legallyenabled through statutory empowerment of the Securities Exchange Boardof India (SEBI) in January 1992. This made possible, for the first time,portfolio inflows from foreign institutional investors without case-by-casegovernment approval, subject to registration with SEBI, and an aggregateceiling of 24 per cent of issued share capital. Portfolio investment isrestricted to foreign institutional investors and high-net-worth individuals.Aggregate percentage ceilings have been steadily liberalized to the presentlevel of 40 per cent. Issue of convertible bonds and shares by Indiancompanies on foreign bourses through American Deposit Receipts (ADRs)and Global Deposit Receipts (GDRs) became permissible for the first timein April 1992, subject to government approval. This was an importantdevelopment, supplementing the opening up to foreign institutional in-vestors, because it enabled companies to reach investors unwilling to getembroiled in the settlements mess on Indian stock markets.

A landmark development was the incorporation in June 1996 of theNational Stock Depository, enabling paperless trading for the first time inIndia. Dematerialization remained optional, but with recognition of theNational Stock Depository by the American Securities Exchange Com-mission in August 1997 as an eligible foreign custodian, it became an optionalmost universally exercised by end-December 1999. Retail investors, otherthan foreign institutional investors, may still opt for the ADR/GDR route,but the premium has declined markedly.

Net equity/investment inflows responded to the lifting of these con-straints and touched a peak of $6 billion in 1996/97 (table 6), largely fuelledby portfolio flows. Portfolio inflows have fluctuated thereafter. Cumulativeinvestment by foreign institutional investors in India from 1993 to mid-2000stood at approximately $12 billion. A phased move, starting in June 1998,towards permission for forward exchange cover for foreign institutionalinvestors has progressed to permission for essentially full cover of exposure(in stages; see appendix A).


ble

6

IND

IA: F

OR

EIG

N IN

VEST

MEN

T/EQ

UIT

Y IN

FLO

WS,

199

0–20

00($

bill

ion)

1990

/91

1991

/92

1992

/93

1993

/94

1994

/95

1995

/96

1996

/97

1997

/98

1998

/99

1999

/00

A.D

irect

Inve

stm

ent

0.10

0.13

0.32

0.59

1.31

2.14

2.82

3.56

2.46

2.16

a.W

ith a

ppro

val (

SIA/

FIPB

)-

0.07

0.22

0.28

0.70

1.25

1.92

2.75

1.82

1.41

b.W

ithou

t app

rova

lR

BI-

-0.

040.

090.

170.

170.

140.

200.

180.

17N

RIa

-0.

060.

050.

220.

440.

730.

760.

600.

460.

57

B.P

ortfo

lio In

vest

men

t0.

010.

000.

243.

573.

822.

753.

311.

83-0

.06

3.03

a.G

DR

sb-

-0.

241.

522.

080.

681.

370.

650.

270.

77b.

FIIs

c0.

010.

000.

002.

051.

742.

071.

951.

18-0

.33

2.26

Tota

l (A

+B)

0.10

0.13

0.56

4.16

5.13

4.89

6.13

5.39

2.40

5.18

Sour

ce:

Res

erve

Ban

k of

Indi

a B

ulle

tin, A

ugus

t 200

0, ta

ble

46. T

here

are

dis

crep

anci

es b

etw

een

the

figur

es h

ere

and

thos

e in

the

BO

P s

tate

men

t.N

ote:

SIA

= S

ecre

taria

t for

Indu

stria

l App

rova

ls; F

IPB

= F

orei

gn In

vest

men

t Pro

mot

ion

Boa

rd; F

II =

Fore

ign

inst

itutio

nal i

nves

tmen

t.a

Incl

udes

acq

uisi

tion

of s

hare

s of

Indi

an c

ompa

nies

by

non-

resi

dent

s un

der s

ectio

n 29

of F

ER

A.

bIn

dian

Cor

pora

te is

sues

of G

loba

l Dep

osito

ry R

ecei

pts

(GD

Rs)

and

Am

eric

an D

epos

itory

Rec

eipt

s (A

DR

s).

cIn

clud

es o

ffsho

re fu

nds.


FDI, on the other hand, has stagnated at around the $2 billion mark,despite the steady expansion of admissible sectors and relaxation ofprocedural requirements. India remains an insignificant destination forglobal FDI flows. Despite automatic FDI entry, possible for the first timein the 1990s, it is entry with approval, for investments crossing sectoralcaps or in sectors requiring licensing, that has remained the dominantcomponent. This may change with the placement in February 2000 of FDIin all sectors, but for a small negative list, in the automatic route.

(ii) Debt

External commercial borrowing is generally understood in India aspertaining to non-trade credit, even though such borrowings may carrysizeable import-earmarked components. There are three main parametersgoverning controls over external commercial borrowing: an aggregate capon such borrowing in any year, with preferences within the cap forinfrastructure and export; restrictions on use for rupee expenditure (i.e.other than foreign currency expenditure on imports), and a ban oninvestment in stock markets and real estate; and minimum maturity limits.

During the mid-1990s surge in capital inflows, the use of externalcommercial borrowing was confined to expenditure as foreign currencyon imports. With the passing of the surge, these rules were gradually easedover time, except at the short-maturity end of the spectrum. Shortermaturities have been discouraged through a lifting of end-use conditionsand approval procedures for longer maturities, subject to caps on theamounts so admissible, which vary directly with the length of the minimummaturity period. Today, end-use restrictions have been lifted altogether,and with that there is a floor maturity of three years. This rules out short-term, non-trade credit altogether.

A peculiar feature of the Indian capital account has been the specialdeposit schemes for non-resident Indians (see box 1), which are accordeda separate row in the balance of payments, distinct from external com-mercial borrowings. Foreign currency deposits carried no exchange riskfor depositors. During the long period until 1997, when deposit rates were

INDIRA RAJARAMAN126

Box 1

DEPOSIT SCHEMES FOR NON-RESIDENT INDIANS

A special deposit schemes for non-resident Indians, NR(E)RA, wasstarted in 1970, during the Bretton Woods era. With the start of amanaged float, a new FCNR(A) scheme was introduced in 1975 withall exchange risk assumed by the RBI. This was replaced in 1993–94by the FCNR(B) scheme, where exchange risk was transferred tocommercial banks. Details on the various schemes and their interestrate structure, are listed below.

Without exchange risk for depositors/banks.FCNR(A) Foreign Currency Non-Resident (Accounts), started

November 1975, closed 15 August 1994.

Without exchange risk for depositors only.FCNR(B) Foreign Currency Non-Resident (Banks), started May

1993.

With exchange risk for depositors.NR(E)RA Non-Resident (External) Rupee Accounts, started

February 1970.NR(NR)RD Non-Resident (Non-Repatriable) Rupee Deposits

(exchange risk only for repatriable interest), startedJune 1992.

NR(S)RA) Non-resident special rupee account with facilitiessimilar to those for resident accounts, introduced inmid-April 1999, but yet to attract inflows.

Interest rates were administratively set untilFCNR(B) 16 April 1997: ceiling rates

22 Oct. 1997: ceiling rate set at LIBOR29 April 1998 : ceiling rate set at

50 basis points > LIBOR for maturity > 1 year; and25 basis points < LIBOR for maturity < 1 year.

Rupee deposits22 April 1992: ceiling rates, usually 100 bp above domestic rates.16 April 1997: rates freed for all but short-term (< 1 yr);

corresponding domestic rates freed in stages betweenOctober 1995 and July 1996.

13 Sept. 1997: rates freed for all (all domestic fixed-term rates> 1 month freed 21 Oct. 1997).


not linked to world rates, they offered clear arbitrage opportunities varyingfrom year to year as a function of the interest rate margin between theadministratively set deposit rate and world deposit rates, with the addi-tional factor of currency risk only in the case of rupee-denominateddeposits. The aggregate net inflows varied accordingly from year to yearin terms of both total and composition.

The interest rate structure on foreign currency deposits marked thefirst time that short-term deposits were formally discouraged, althoughthe earlier administered-rate regime is thought to have incorporated asimilar tilt in the rate structure. In October 1999, the minimum maturity offoreign currency, non-resident Indian deposits (FCNRB) was lengthenedto one year, from a floor maturity which had all along been set at six months,thus doing away with short-term deposits altogether. What is important tonote is that these policy nudges away from short-term deposits are recent,a clear effect of the 1997 East Asian crisis. The RBI has set up a SteeringCommittee and a Technical Group to work on a risk management modelfor sovereign external liability management in India.

The total stock of external debt has been well under $100 billion inthe 1990s (table 7). Differencing these stock data does not yield annualflows comparable to those in the balance of payments table, however,because of the valuation effect. Additionally, non-resident Indians’ non-repatriable rupee deposits are excluded from external debt stock figures,quite justifiably, perhaps, but this does pose problems of comparabilitywith the balance-of-payments flow data.

Short-term trade credits, and short-term deposits by non-residentIndians (until the October 1999 decision imposing a minimum maturity ofone year) together constitute aggregate external short-term liabilities. Short-term debt in Indian official figures is classified by original, not residual,maturity.

In stock terms, short-term liabilities, aggregating across (suppliers’)trade credits and non-resident Indian deposits, fell from $7.5 billion atend-March 1991 to $4.1 billion at end-March 2000, and in percentage terms,from around 10 per cent to around 4 per cent of total external liabilities.


ble

7

IND

IA: E

XTER

NA

L D

EBT

STO

CK

, 199

1–20

00($

bill

ion)

Long

-term

com

mer

cial

bor

row

ing

Trad

e cr

edits

aN

RI d

epos

itsb

Shor

t-ter

m

Tota

lLo

ans

Bond

scAg

greg

ate

Shor

t-ter

mSh

ort-t

erm

Aggr

egat

eSh

ort-t

erm

Shor

t-ter

mAg

greg

ate

Tota

lEn

d-M

arch

(Per

cen

t)(P

er c

ent)

(Per

cen

t)

1991

83.8

06.

833.

209.

104.

8052

.75

12.9

52.

7421

.16

7.54

9.00

1992

85.2

86.

724.

878.

224.

2351

.46

12.9

22.

8421

.98

7.07

8.29

1993

90.0

26.

495.

157.

282.

9640

.66

14.5

23.

3823

.28

6.34

7.04

1994

92.6

96.

316.

056.

991.

7925

.61

14.4

91.

8312

.63

3.62

3.91

1995

99.0

06.

666.

338.

621.

9923

.09

15.7

63.

3821

.45

5.37

5.42

1996

93.7

37.

236.

647.

532.

1528

.55

14.8

93.

8826

.06

6.03

6.43

1997

93.4

78.

535.

818.

812.

9533

.48

14.7

83.

7725

.51

6.72

7.19

1998

93.5

39.

987.

019.

382.

8530

.40

14.1

02.

1915

.53

5.05

5.40

1999

97.6

710

.34

10.7

58.

982.

1924

.39

14.5

42.

2015

.13

4.39

4.49

2000

98.4

49.

959.

429.

212.

5627

.80

16.0

61.

489.

224.

044.

10

Sour

ce:

Ibid

., ta

ble

137,

cor

rect

ed w

ith re

fere

nce

to R

BI,

Ann

ual R

epor

t 200

0, a

ppen

dix

tabl

e V

I.9.

aU

nlik

e th

e B

OP

tabl

e 12

2, e

xter

nal d

ebt s

tock

s lis

t (m

&lt)

trad

e cr

edits

sep

arat

ely

from

com

mer

cial

bor

row

ing.

bN

on-r

esid

ent I

ndia

n de

posi

ts h

ere

excl

ude

non-

repa

triab

le ru

pee

deno

min

ated

dep

osits

. Thu

s, th

e flo

ws

obta

ined

by

diffe

renc

ing

the

stoc

k fig

ures

her

e w

illno

t cor

resp

ond

to b

alan

ce o

f pay

men

ts fl

ows,

qui

te a

side

from

the

curr

ency

val

uatio

n ef

fect

.c

Bon

ds a

re a

ll se

curit

ized

bor

row

ings

incl

udin

g ne

t inv

estm

ent b

y 10

0 pe

r cen

t deb

t fun

ds o

f for

eign

inst

itutio

nal i

nves

tors

, and

incl

udin

g th

e R

esur

gent

Indi

aB

onds

issu

e of

199

8.


The decline was sharpest and most steady in trade credits. Short-term non-resident Indian deposits rose to touch a peak of nearly $ 4 billion beforefalling after 1997, in a clear response to the formal linking of deposit ratesto the London Inter-Bank Offered Rate (LIBOR) established after that (thisshows that whatever attempts were made to curb short maturities beforethat were not very successful). There are no balance-of-payments flowdata on short-term non-resident Indian deposits (table 3). There are onlythe stock data which show a sharp decline after 1997, inclusive of thevaluation effect.

Short-term trade credit inflows as recorded in the balance of payments(table 3) were never exceptionally large, of the order of $1 billion even inthe pre-crisis year, 1990/91. In subsequent years these declined to negativeflows, again only of the order of $1 billion at most in any year. However,these flows are complete only in respect of suppliers’ credits. Short-termcredits of less than six months’ maturity through banking channels, backedby letters of credit, do not appear even in the residual “other capital” row,because commercial banks are not required to report trade credits of lessthan six months’ original maturity, for which no foreign exchange approvalis required.

The Bank for International Settlements (BIS) publishes external debtstock figures which do include short-term credits through banking channels.Despite this, total trade credits reported for India by the BIS are below theIndian figure. There are other areas of incomplete coverage in the BISdata.2 This is unfortunate, since the BIS is the only source on short-termdebt based on residual maturities. Thus the Indian official figures offer theonly series available with full coverage of all debt, and, within the param-eters of their limitations, they show a halving in the proportion of short-termdebt by original maturity over the decade.

INDIRA RAJARAMAN130

E. Returning flight capital

It is clear that there has been return of flight capital in the 1990s on amajor scale, motivated by the progressive liberalization of the currentaccount (including the legalization of gold imports already referred to),leading up to the August 1994 acceptance of IMF Article VIII status. Bymaking possible legitimate purchases of foreign currency and therebyreducing the need for external holdings of foreign currency, current accountliberalization has thus largely financed itself.

The fact of returning flight capital is acknowledged, but there aredifferences of opinion on the particular avenues used. One popular bodyof opinion holds that returning flight capital in India has been largelythrough export overinvoicing (export income throughout the 1990s wasnon-taxable; a first move towards phased taxation of export income wasbegun in April 2000). Somewhat confusingly, the impact of exportoverinvoicing is thought to be statistically evident, not so much in thecurrent account itself, where export figures are entered as declared atcustoms, but in the capital account in the “other capital” row, where foreigncurrency receipts in excess of customs declarations are accommodated.The argument against that holds that the errors and omissions componentof “other capital” is not large enough to suggest export overinvoicing on amassive scale. Another possible channel is through gold imports, althoughit is possible that transfers financing these have only moved to the formalfrom the informal channel, but continue to originate from earnings of non-residents rather than from flight capital of resident Indians.

Non-resident Indian deposits are among the capital account channelsavailable for returning flight capital. The inflows into the non-repatriablerupee schemes in particular are thought to be evidence of flight capitalreturn. The equity issues abroad (ADRs and GDRs) by Indian corporatesand bond issues, most recently Resurgent India Bonds in late 1998, whichraised about $4 billion, offer other avenues for flight capital return.

No official estimates are available of the stock of flight capital (Reddy,2000, denied the existence of it) nor, needless to say, of inflows of returning


capital since 1991. A popular estimate is $9–10 billion a year, but this isnot rigorously underpinned.

F. The foreign exchange market

The Indian foreign exchange market remains thin, as acknowledgedin the RBI Annual Report, 2000. The price of the rupee has been market-determined ever since March 1993, with no official control over trends, asdistinct from episodes of volatility during which the RBI does activelyintervene. The rupee depreciated nominally, by around 7 per cent annuallyover the period 1995–2000 (see table 8). Although India escaped unscathedduring the Asian crisis of 1997, insulated from contagion effects by thelow exposure of foreign portfolio investors to the country, there have inrecent years been many episodes of volatility – periods of sharp downwardpressure on the rupee – which the RBI has actively intervened to correct.During the period 1999–2000, there were three such episodes: in end-May1999, end-August 1999, and again in end-May 2000.

In correcting these, the RBI’s intervention strategy relied essentiallyon aligning domestic short-term rates, in particular in the call-moneymarket, with interest rates implicit in forward premiums, by using its leversof control over the liquidity in the system so as to remove incentives toarbitrage. The standard levers of control used by central banks anywhere,the bank rate and the cash reserve ratio to be maintained by banks with thecentral bank, have been augmented with a liquidity adjustment facility(LAF), implemented starting June 2000 in a phased manner, wherebyliquidity is injected/removed from the system through reverse repo/repoauctions. The recommendations of the two Narasimham Committees haveserved as templates for financial sector reform in India, and the LAF wasintroduced on the recommendation of the second Narasimham Committee(Government of India, 1998; 1991).

The LAF will eventually completely replace the present set of mecha-nisms involving collateralized short-term lending facilities and a fixed raterepo provision. LAF auctions are conducted daily, and in August 2000 the


ble

8

IND

IA: E

XCH

AN

GE

RAT

ES A

ND

FIN

AN

CIA

L SE

CTO

R IN

DIC

ATO

RS,

199

0–20

00

1990

/91

1991

/92

1992

/93

1993

/94

1994

/95

1995

/96

1996

/97

1997

/98

1998

/99

1999

/00

Exc

hang

e ra

te (R

s/$)

End

per

iod

19.6

431

.23

31.2

431

.37

31.5

34.3

535

.92

39.5

42.4

443

.62

Ave

rage

17.9

424

.47

30.6

531

.37

31.4

33.4

535

.50

37.1

642

.07

42.6

5

Com

mer

cial

ban

k in

tere

st ra

tes

(end

-per

iod)

(per

cen

t)

Dep

osit

rate

(1 y

r.)9.

012

.011

.010

.011

.012

.011

.010

.59.

07.

5LI

BO

R (1

yr.)

8.5

6.3

4.2

3.6

5.6

6.2

5.8

6.1

5.5

5.7

(Dep

. rat

e –

LIB

OR

)0.

55.

76.

86.

45.

45.

85.

24.

43.

51.

8In

flatio

n ra

te13

.413

.66.

88.

39.

98.

210

.27.

29.

05.

0R

eal d

epos

it ra

te-4

.4-1

.64.

21.

71.

13.

80.

83.

30.

02.

5

Lend

ing

rate

s:M

in.

16.0

19.0

17.0

14.0

....

....

....

PLR

....

....

15.0

16.5

14.8

14.0

12.5

12.3

Sour

ce:

RB

I, H

andb

ook

of S

tatis

tics

on In

dian

Eco

nom

y, 1

999,

tabl

es 1

27, 1

56; s

uppl

emen

ted

by R

BI,

Ann

ual R

epor

t 200

0, a

ppen

dix

tabl

e I.2

and

RB

I, M

onth

lyB

ulle

tin; G

over

nmen

t of I

ndia

, Eco

nom

ic S

urve

y, 1

999–

2000

, tab

le 5

.3; a

nd IM

F, In

tern

atio

nal F

inan

cial

Sta

tistic

s fo

r LIB

OR

; ave

rage

Jan

./May

for 2

000.

Not

e:D

epos

it ra

tes

star

ting

from

199

5/96

, and

lend

ing

rate

s st

artin

g fro

m 1

994/

95 w

ere

freed

; the

PLR

cam

e in

to b

eing

effe

ctiv

e O

ctob

er 1

994,

with

the

rate

s he

reth

e of

ficia

lly re

porte

d av

erag

es a

cros

s th

e fiv

e bi

gges

t pub

lic s

ecto

r ban

ks. E

ffect

ive

29 O

ctob

er 1

999,

ban

ks w

ere

freed

from

ben

chm

arki

ng to

the

PLR

for

certa

in c

ateg

orie

s of

adv

ance

s. B

anks

are

requ

ired

to p

resc

ribe

a m

axim

um s

prea

d ov

er P

LR s

ince

199

6/97

, whi

ch th

ey a

re fr

ee to

det

erm

ine.

Infla

tion

rate

sar

e fro

m y

ear/e

nd v

alue

s of

the

CP

I for

urb

an n

on/m

anua

l wor

kers

. LIB

OR

rate

s pe

rtain

to c

alen

dar y

ear a

vera

ges

as re

porte

d in

the

sour

ce, o

verla

ppin

g w

ithth

e In

dian

fisc

al y

ear f

or th

ree

of th

e fo

ur q

uarte

rs.


tenor of the repos auctioned was extended from one day up to a maximumof seven days. At the same time, the players authorized to participate inrepo auctions (i.e. to borrow from the RBI, and thus to get access to short-term liquid funds) have been steadily expanded to include non-bank entities,after a ban for some years on repos as a consequence of a scam in 1992involving the misuse of funds obtained through repos. The set of approvedmarket makers in both the primary and secondary markets in governmentsecurities, termed primary dealers, now numbers 15. An attempt is beingmade to impose internal control systems and capital adequacy requirementson non-bank primary dealers so as to place them on par with bank primarydealers.

Even with all the widening in levers of control over liquidity in thesystem, the RBI has had to resort to ad hoc direct tampering with interestrates on exports (export credit rates are still controlled by the RBI), andsurcharges on import finance (which are not controlled, hence the use ofsurcharges) in order to achieve targeted outcomes in the foreign exchangemarket.

The abrupt alteration in these during a volatility episode in mid-2000was seen as a hark back to an earlier era of instability in policy parameters.There was particular resentment over the halving in August 2000 of bothbalances in, and permissible accretions to, the exchange earners’ foreigncurrency (EEFC) accounts that exporters could maintain abroad. The EEFCScheme introduced in 1992 allowed exporters for the first time to retain aprescribed percentage of their receipts in foreign exchange with anauthorized dealer in India. The August 2000 ruling was reversed in October,with, however, retention of the new constraint introduced with that ruling,confining EEFC holdings to non-interest-bearing current accounts.

A forward market for foreign exchange now functions. The devel-opment of the forward market was given a major impetus with first-timepermission in April 1997 for forward contracts without documentaryevidence of underlying exposure, and beyond six months, subject, however,to a declaration of exposure and fairly stringent documentation. At thesame time, the case-by-case approval of currency swaps was replaced bypermission for authorized dealers to operate “swap books” within their

INDIRA RAJARAMAN134

open position limits. Forward cover for portfolio investments has beenpermitted starting in June 1998, extending in April 1999 to essentially thefull extent of exposure (appendix A).

G. Financial sector reform

Starting from a pre-reform scenario in which all deposit and lendingrates were prescribed, the interest rate deregulation introduced in October1994 has left only a single vestigial deposit rate still set by the RBI onsavings (non-term) deposits. Banks are now free even to offer floatingrates on term deposits. However, there remain many more controls onlending rates, on small loans (up to Rs 200,000) and subsidized, differentialrate of interest (DRI) loans to deprived sections, and on export credit, whichare prescribed at flat rates or capped. Lending rates on other loans arecapped at a maximum spread over the prime lending rate (PLR), with eachbank free to set both the benchmark PLR and the spread. Since October1997 banks have been free to set term-linked PLRs, and since October1999 they have also been freed from PLR benchmarking for certaincategories of advances.

Interest rate deregulation found banks in India with poor in-houserisk-assessment abilities. Indeed, the requirement of a maximum spread(as set by each bank) over PLR was introduced because banks’ inability toassess risk for sub-prime borrowers led initially to very large margins overthe PLR, far in excess of the underlying credit risk. The only formal liquidityrequirements in place (table 9) are the cash reserve ratio (CRR) and thestatutory liquidity ratio (SLR). During the pre-crisis period, the formerwas set at 15 per cent, while the latter, which provides a captive market forgovernment securities to finance fiscal deficits, was set at 38.5 per cent.Both the CRR and SLR have been brought down in tandem with interestrate deregulation, although banks continue to subscribe to risk-freegovernment securities well in excess of the present SLR of 25 per centbecause of continued in-house lacunae in credit and market risk assessmentcapabilities. The scrapping of CRR and SLR on inter-bank borrowings hasreduced friction in the call-money market and has led to the Mumbai Inter-


Bank Offered Rate (MIBOR), which could develop into an effectivebenchmark (see also Williamson and Mahar, 1998).

The RBI issued a detailed set of guidelines for putting in place anasset-liability management (ALM) system for commercial banks inFebruary 1999, and for other financial institutions in January 2000. Theguidelines are equipped with periodic reporting requirements to the RBI,albeit without statutory conformity requirements or penal clauses. Mean-while, latitude in respect of fund deployment continues to widen. Sincethe latest liberalization in early September 2000, banks have been permitted5 per cent of total outstanding advances in equity and equity-linkedinstruments in the form of shares, convertible debentures, and units ofmutual funds. This is expected to deepen equity markets, reduce the domi-nance of external portfolio investors and lead to a revival of slow-growthstocks.

Table 9

INDIA: MONETARY CONTROL LEVERS, 1990–2000(Per cent end period)

Bank rate Cash reserve ratio (CRR) Statutory liquidity ratio

1990/91 10.0 15.0 38.51991/92 12.0 15.0 38.5a

1992/93 12.0 15.0 37.3a

1993/94 12.0 14.0 34.8a

1994/95 12.0 15.0 31.5a

1995/96 12.0 14.0 31.5a

1996/97 12.0 10.0 31.5a

1997/98 10.5 10.3 25.01998/99 8.0 10.5 25.01999/2000 8.0 9.0 25.0

Source: RBI, Handbook of Statistics on Indian Economy, 1999, table 36, supplemented by RBI, AnnualReport 2000.

Note: For change after 1 April 2000, see box in text.a These were on the stock of net liabilities on specified dates, with a reduced rate on incremental

liabilities beyond that date.

INDIRA RAJARAMAN136

A large number of initiatives have been undertaken to reduce frictionin the banking system, in payments settlements for both domestic and cross-border settlement. This will, among other beneficial effects, have a loweringeffect over time on rates at the short end of the market for liquidity. Amajor advancement towards enabling hedging of interest rate risks wasmade with the issue of guidelines in July 1999 for two rupee derivativeinstruments, forward rate agreements (FRAs) and interest rate swaps (IRS),permitting even market making participation without underlying exposure,but they are confined to plain vanilla contracts without explicit or implicitoption features such as caps/floors/collars.

Permission to hedge commodity price exposures through participationin futures or options contracts in international exchanges is now grantedby the RBI on a case-by-case basis. The recognition of the loss suffered in2000 on account of unhedged imports of oil at a time of rising internationaloil prices is likely to speed up the pace of deregulation on this front. Majorobstacles, however, are the large public sector presence in internationalcommodity imports and internal constraints on their functioning on accountof audit and vigilance regulations.

The traditional division in India between banks, as providers ofworking capital, and three term-lending financial institutions, as providersof project finance, was removed some years ago. The financial institutionsare gradually being brought under the regulatory mantle of the RBI. At thesmaller end of the size spectrum, non-bank financial institutions, have beenbrought under prudential control for the first time through the ReserveBank Amendment Act of 1997, with a minimum capital requirement forregistration and a higher requirement for the entitlement to accept freshpublic deposits.

The share of non-performing assets (NPAs) in total bank loans hasbeen a cause of worry for some years. Long time series on NPAs are notpossible because there has been a phased alteration of the definition of NPAscompleted only in 1994/95 (see Rajaraman, Bhaumik and Bhatia,1999).The most recent figures available relate to end-March 2000, and show agross NPA percentage of 12.8 per cent, and a corresponding net figure of6.8 per cent (see RBI, 2000). Although lower than the figures recorded at


end-March 1997 of 15.7 per cent (gross) and 8.1 per cent (net), these arehigh figures by any standards, and clearly far removed from the target of5 per cent gross set by the Tarapore Committee for end-March 2000 as aprior requirement for full capital account convertibility (see RBI, 1997). Aparticularly interesting feature of NPAs in India is the enormous cross-sectional variation between banks as a function of vintage, ownership andregion of operation (for regression results explanatory of cross-sectionalNPA variations for 1996/97, see Rajaraman, Bhaumik and Bhatia, 1999).

A menu of options for the NPA problem has been offered to the bankingsystem through Debt Recovery Tribunals and Settlement AdvisoryCommittees. But these are hampered by an inadequate legal infrastructure(see RBI, Report on Currency and Finance, 1999–2000). The Governmentof India Budget for 2000/01 in February 2000 announced plans to establisha Financial Restructuring Authority to oversee recapitalization of weakbanks in the system.

The minimum capital to risk assets ratio has been raised from 8 to9 per cent effective as of 31 March 2000. The Government has announcedits intention to dilute its shareholding in public sector banks, whichdominate the banking sector, to 33 per cent, and banks are being encouragedto raise the requisite capital through the market rather than through strategictie-ups.

The Insurance Regulatory and Development Authority Act was passedin 1999, whereby banks and non-bank financial companies are permittedto enter the insurance market, breaking a government monopoly in placefor many decades in both life and non-life segments.

H. Conclusions

It is difficult to disentangle the essence of India’s approach to capitalaccount management through the thicket of liberalization notifications thathave been issued in a steady stream since the onset of the opening up ofthe economy in July 1991, bearing the Indian policy hallmark of gradual

INDIRA RAJARAMAN138

incrementalism. What is clearly in evidence, however, is the mono-tonicity of the direction of movement towards liberalization of cross-bordercapital flows.

The capital account today in India carries convertibility for non-residents but not for residents. Capital inflows, whether in the form of FDIor portfolio investment, are granted full freedom of repatriation. Theredoes exist a non-repatriable rupee deposit scheme for non-resident Indians,but it is an option exercised by the depositor, co-existing with repatriableschemes, and is clearly an invitation to returning flight capital. Residents,on the other hand, are not permitted to hold foreign currency accounts,except for exporters and other exchange earners in EEFC accounts. Outwardportfolio investment flows from non-corporate residents are prohibited.Outward investments by corporates through share swaps are permitted underthe new Foreign Exchange Management Act, effective from 1 June 2000,without approval subject to sectoral caps, and with approval otherwise.Under the new Act, contraventions of the ban on outward capital flows aretreated as a civil, rather than criminal, misdemeanour, a major alterationin the legal regime governing the capital account.

Capital account liberalization in India has been slow and cautious.The 1997 Report of the RBI (Tarapore) Committee recommended a three-year phased move to full convertibility by 2000 subject to achievement ofmacro targets (see appendix A) for the (overall) fiscal deficit, inflation,and the debt-service ratio, apart from targets for financial sector indicators.Of these, the fiscal deficit target of 3.5 per cent of GDP in 1999/2000 isnowhere near actuals for the year, which were 5.6 per cent for the centralGovernment alone, and 9.9 per cent combining central and state govern-ments.

The institutional changes needed to improve regulation of the stockmarket without strangulating government control were introduced at theoutset, but arrangements to dematerialize transactions in the share marketwere slow to follow. It is only as recently as April 1999 that the stocktransactions included in the major indices were fully dematerialized. Thisreduction of friction in trading sets the stage for the development of stockmarket derivatives.


The reform of the economy was motivated by an external liquiditycrisis, with reserves at a low of $1 billion. At end-March 2000, reservesstood at $38 billion, a first attestation of the success of the reform process.

The introduction of current account convertibility in 1994 indisputablybrought flight capital back into the country, but there is some dispute as towhether the avenue of return was the current account (through the financingof newly- permissible gold imports and through export overinvoicing), orthe capital account (through the special deposit schemes and periodicexternal bond issues open to non-resident Indians). Whether fuelled byreturning flight capital or not, the non-governmental flows on capitalaccount amounted in the mid-1990s to well in excess of 100 per cent of thecurrent account balance, so that until as recently as 1997 the capital accountmanagement problem was one of too much rather than too little.

FDI was not among the channels that contributed to the mid-decadesurge. Notwithstanding steady liberalization in terms of required approvalsand export obligations attached to dividend repatriation, FDI has been slowto respond, hampered, not as earlier by national entry obstacles, but by thecontinuing structural rigidities of the Indian economy, and by an absenceof coherence between policy at the national level and procedural obstaclesat the local level (Rajaraman, 1997). It was portfolio inflows, from bothforeign institutional investors and external ADR/GDR share capital issues,that contributed to the mid-decade surge, responding to full repatriabilitywith no lock-in periods.

The problem of excess capital inflows in the mid-1990s was tackledby restricting the end-use of equity capital raised through ADR/GDR issuesand external commercial borrowings to foreign currency expenditure onimports, and by prohibiting use for rupee expenditure within the country(these stalled inflows were recorded as net outflows in the “other capital”row). As a technique of capital account management it was far superior toa more market-based approach of free entry with sterilization, anddemonstrates the effectiveness of capital barriers in handling imbalanceson the capital account. As the capital inflow surge ceased, the end-userestrictions were lifted. Today there are no such restrictions on ADR/GDR

INDIRA RAJARAMAN140

equity capital or on debt inflows above prescribed floor maturities. Thesefloor maturities on non-trade external commercial borrowings have beeneffectively used to contain non-trade short-term debt.

A peculiar feature of external debt inflows into India is the menu ofdeposit schemes offered to non-resident Indians, which during the 1990shave seen a decisive shift towards rupee-denominated, and, within rupee-denominated, non-repatriable deposits (introduced in 1992). With non-repatriable rupee deposits, only the interest is repatriable (with currencyrisk). The foreign currency non-resident Indian deposit schemes lost theirarbitrage advantage when the ceiling on deposit rates in 1997 was set atLIBOR. The first formal discrimination against short-term inflows withinthe non-resident Indian scheme in April 1998 was clearly motivated by theAsian currency crisis of late 1997. The impact on the short-term share ofnon-resident Indian deposits is clearly visible. With the raising of theminimum maturity on such deposits to one year in October 1999, short-term non-resident Indian deposits have been done away with altogether.

On non-repatriable, non-resident Indian rupee deposits, the principalpayable at maturity is contra-entered in the current account as privatetransfers. This accounting artefact has been one of the two major causes ofthe extraordinary increase in private transfers to annual inflows of around$15 billion today. The second is of course the legalization of gold imports,which has resulted in the restoration of private transfers to the formalchannel from informal channels, not only as contra-entries in the firstinstance to legal gold imports, but also as secondary effects of the reducedattractiveness of exchange rates through the informal hawala channel.

The only long-term series available at the time of writing is the officialIndian series from which it seems possible to conclude that there has beenan indisputable fall in short-term trade credit since the early 1990s. Theissue of what has happened in the last few years to short-term trade creditsremains unresolved. BIS figures of short-term debt by residual maturityare uncomfortably inconsistent with the Indian figure by original maturityin terms of both level and direction of movement, but since the BIS figuresexhibit evidence of incompleteness in respect of both the total debt stockand total trade credits, they are not immediately acceptable as the correct


source on the direction of movement in recent years in respect of short-term trade credit.

Not surprisingly, the market for foreign exchange remains thin, withperiods of volatility that have seen RBI intervention through not merelythe use of monetary levers of control over liquidity in the system, but alsoad hoc tampering with rates charged on export credit, which remain underadministrative control, along with surcharges on import finance, and ahalving (albeit short-lived) of balances in, and permissible accretions to,external (EEFC) accounts, as recently as August 2000. While none of thesecan even remotely be characterized as reversing the direction of themovement towards capital account liberalization, the sudden reversal ofpolicy parameters, for example on EEFC accounts, is seen as a throwbackto the pre-reform era of policy unpredictability.

If there should be another and more pronounced surge in foreign capitalinflows, the process of liberalizing capital outflows may be carried furtherforward, beginning with corporate investment, and proceeding eventuallyto non-corporate portfolio outflows. In the interim, while the reduction infinancial market friction is in itself an encouragement to capital inflows, itis the underlying real economy, and the continued fiscal, infrastructureand labour market bottlenecks that are in need of urgent reform.

III. MALAYSIA

A. Introduction

Malaysia prior to the Asian crisis of 1997 had had a long history ofopenness on both current and capital accounts, unique in the developingworld, and ahead even of some developed countries. Current accountconvertibility under Article VIII of the IMF was accepted as far back asNovember 1968, and the Malaysian ringgit was among the first currenciesto go into a float in May 1973, after the collapse of the Bretton Woodssystem (see appendix B). With the resulting need for hedging instruments,

INDIRA RAJARAMAN142

the greater availability of these in more sophisticated offshore financialmarkets led to the development of an offshore ringgit market, locatedprincipally in Singapore. Malaysian exports and imports were ringgit-denominated, and the burden of hedging was transferred to non-residentcounterparties in the offshore market.

Capital flows were liberalized in three major moves: in 1973, alongwith the ringgit float; in 1987, as part of a policy package to pull out of themid-1980s recession; and in 1994, as part of a package of measures designedto handle excessive capital inflows. The Malaysian response to downwardpressures on the exchange value of the ringgit during the Asian crisis of1997 reversed this traditional openness. The long history of a floatingexchange rate was abruptly terminated by a ringgit peg at 3.8 ringgit to thedollar on 2 September 1998, a day after the announcement of a package offinancial policy measures. These measures were designed to kill the offshoreringgit market and to prevent all access by non-residents to ringgit, as wellas to obstruct outward speculative flows of capital from residents. Theyalso aimed to drive a further wedge between the foreign exchange marketand the domestic imperative to lower interest rates which had been drivenup during the crisis by the imposition of a 12-month stay on portfoliooutflows, though only in respect of the principal. Meanwhile, interest anddividends remained freely repatriable and the current account remainedfree, apart from checks on porosity by requiring repatriation of exportproceeds within six months from the date of export. There was also a limiton exports of foreign currency by residents, which functioned as a cap oninvisible outflows.

As a package, the Malaysian retreat from capital account convertibilityhas to be among the major milestones in recent economic history. Itchallenged the prevailing economic orthodoxy, and was orthogonal to therecovery packages imposed by the multilateral financial institutions onother East Asian countries similarly afflicted. By not curtailing capitalaccount convertibility, and therefore not severing the link between monetarypolicy and exchange markets, the recovery measures meant tight moneyand high interest rates, with devastating consequences for the domesticreal economy in these neighbouring countries.


In what follows, capital account management in Malaysia prior to1997 is dealt with in section III.B, and the crisis itself in section III.C. Thecapital controls of 1998/99 are dealt with in section III.D, and the post-crisis financial sector reform measures in section III.E. Section III.Fconcludes.

B. Capital account management prior to 1997

Table 10 shows the Malaysian balance of payments in the 1990s. Thetwo most striking features of the Malaysian current account were: (i) theconsistently positive, though small (except in 1998, a recession year),balances on merchandise trade, with large export earnings at $70–80 billiontowards the end of the decade matched by an equivalently large importbill; and (ii) steady outflows on invisibles, now close to $10 billion annually.The current account has thus always been in deficit (except in 1998). Infact, the current account deficit averaged around 7 per cent of GDP in thethree years preceding the crisis.

Buoyant capital inflows more than compensated for the deficit oncurrent account. Indeed, in the first half of the 1990s, non-official capitalinflows exceeded the current account deficit by a factor of 4.5, and foreignreserves jumped from $11 billion at end-1991 to $28 billion at end-1993(table 11), as the central bank, Bank Negara Malaysia (BNM), sought tocontain the appreciation of the ringgit. The short-term component of inflowswas especially large, at around $5 billion in both 1992 and 1993. Inwardcapital controls were introduced in 1994 but withdrawn within the year(appendix B). These restrictions, which covered external borrowing,portfolio investment in Malaysian securities, and forward and swaptransactions by banks, succeeded in reducing non-official capital inflows,and changed the direction of short-term flows from inflows of $5.4 billionin 1993 to outflows of $3.2 billion in 1994 (table 10). The temporary capitalcontrols of 1993/94 were a minor blip, however, in what was, untilSeptember 1998, an unfettered regime for external capital inflows andrelatively few constraints on capital outflows. Indeed, the response to theexcess inflows of 1994 included a further relaxation on outward portfolioinvestment in February 1994.


ble

10

MA

LAYS

IA’S

OVE

RA

LL B

ALA

NC

E O

F PA

YMEN

TS, 1

990–

1998

($ b

illio

n)

Item

1990

1991

1992

1993

1994

1995

1996

1997

1998

I.M

erch

. tra

de b

alan

ce (A

-B)

2.62

0.53

3.38

3.20

1.70

0.04

4.01

3.65

17.5

9A

.E

xpor

ts, f

.o.b

.28

.64

33.5

439

.62

45.9

956

.61

71.5

676

.86

77.3

971

.87

B.

Impo

rts, c

.i.f.

26.0

233

.01

36.2

442

.79

54.9

171

.53

72.8

573

.74

54.2

8

II. I

nvis

ible

s, n

et

-3.5

4 -4

.76

-5.5

9 -6

.28

-7.3

3 -8

.67

-8.4

7 -9

.28

-8.2

1

III.

Cur

rent

Acc

ount

(I+I

I) -0

.92

-4.2

3 -2

.21

-3.0

8 -5

.63

-8.6

3 -4

.46

-5.6

2 9.

38/G

DP

(per

cen

t) -(

2.09

) -(

8.62

) -(

3.73

) -(

4.60

) -(

7.56

) -(

9.73

) -(

4.42

) -(

5.61

) (1

2.93

)

IV.

Cap

ital a

ccou

nt (A

to D

), ne

t2.

905.

488.

7814

.44

2.48

6.88

6.94

1.75

0.89

A.

Priv

ate

long

-term

cap

ital

2.33

4.00

5.18

5.01

4.12

4.17

5.08

5.14

2.16

B.

Offi

cial

long

-term

cap

ital

-1.0

5-0

.24

-1.1

30.

380.

332.

450.

301.

650.

54C

.P

rivat

e sh

ort-t

erm

cap

ital

0.50

1.87

4.69

5.41

-3.2

31.

014.

10-4

.59

-5.2

6D

.E

rror

s an

d om

issi

ons

1.12

-0.1

40.

033.

641.

27-0

.76

-2.5

3-0

.45

3.44

V.O

vera

ll B

alan

ce (I

II+IV

)1.

981.

256.

5711

.36

-3.1

5-1

.76

2.48

-3.8

710

.27

VI.

Res

erve

s an

d re

late

d ite

ms

-1.9

8-1

.25

-6.5

7-1

1.36

3.15

1.76

-2.4

83.

87-1

0.27

IV [A

+C+D

]/III

(per

cen

t)(4

30)

(135

)(4

48)

(456

)(3

8)(5

1)(1

48)

(2)

..

Sour

ce:

Ban

k N

egar

a M

alay

sia

(BN

M),

Mon

thly

Sta

tistic

al B

ulle

tin, J

une

2000

, tab

les

V.6

, VI.3

and

VIII

.1. F

igur

es fo

r 199

9 w

ere

avai

labl

e on

ly u

pto

Q3.

Not

e:Fi

gure

s in

ring

git c

onve

rted

to U

S d

olla

rs u

sing

per

iod

aver

age

exch

ange

rate

s (s

ee ta

ble

12).

The

bala

nce

of p

aym

ents

figu

res

for p

rivat

e lo

ng-te

rm c

apita

lin

clud

e ex

tern

al b

orro

win

gs a

nd a

ccru

ed e

arni

ngs

on F

DI;

and

are

net o

f FD

I out

flow

s. T

here

nee

d th

eref

ore

be n

o ab

solu

te n

or d

irect

iona

l cor

resp

onde

nce

betw

een

thes

e fig

ures

and

the

sum

of g

ross

FD

I inf

low

s an

d ne

t por

tfolio

inflo

ws

in ta

ble

2. F

igur

es in

par

enth

eses

in th

e la

st ro

w p

rese

nt th

e ra

tio o

f tot

alpr

ivat

e ca

pita

l inf

low

s to

the

curr

ent a

ccou

nt d

efic

it in

per

cent

age

term

s; th

ere

is th

eref

ore

no fi

gure

for 1

998

whi

ch h

ad a

cur

rent

acc

ount

sur

plus

.


The following constituted the configuration of capital account manage-ment prior to the East Asian crisis of 1997:

• Portfolio capital inflows by non-residents were freely permissible intoall types of Malaysian financial instruments (bonds, equities, bankdeposits, and all money market and derivative instruments).

• Primary issue of securities, whether abroad by residents or in Malaysiaby non-residents, required prior approval.

• Portfolio capital outflows were unrestricted for individual residentsand for corporates with no domestic borrowing; for those withdomestic borrowing, prior approval was needed for remittances inexcess of 10 million ringgit per corporate group per year.

Table 11

MALAYSIA: EXTERNAL RESERVES, 1990–2000($ billion)

International reserves

End Special IMF Gold and foreign Gross internationalof period drawing rights drawing rights exchange reserves

1990 0.20 0.23 9.59 10.021991 0.21 0.26 10.72 11.191992 0.11 0.33 17.68 18.121993 0.12 0.31 27.88 28.311994 0.14 0.40 26.13 26.661995 0.15 0.68 24.28 25.111996 0.17 0.69 26.85 27.701997 0.12 0.42 14.67 15.211998 0.21 0.63 25.33 26.171999 0.09 0.83 29.94 30.862000 (June) 0.09 0.89 33.01 34.00

Source: BNM (1999), tables II.6 and II.7, updated with BNM, Monthly Statistical Bulletin, June 2000.Note: IMF drawing rights refer to Malaysia’s quota in the International Monetary Fund, less IMF holdings

of Malaysian currency.

INDIRA RAJARAMAN146

• FDI inflows were not merely freely permissible, although with priorapproval in a few sectors, but were welcomed with tax incentives andother kinds of incentives.

• FDI outflows were freely permissible.

• External borrowing was unrestricted for authorized dealers, includinginvestment banks, with no restrictions on on-lending to residents ornon-residents, subject to prudential limits on net open positions. Directexternal borrowing by residents was subject to approval abovesanctioned limits.

Overall, the package added up to a well-structured liberalization ofcapital flows, with limits and checks at critically necessary points, such asat the point of primary issue of securities, both inward and outward; ondomestically leveraged large outward portfolio flows; and on direct externalborrowing by non-institutional residents (see also Cole, Scott andWellans,1995).

The second major push in 1987 towards liberalizing the external capitalflows regime was followed by financial sector reform and strengtheningof prudential controls during the period 1989–1995. New legislation in1989 strengthened regulatory and prudential provisions relating to loanclassification, provisioning and disclosure, capital adequacy, and exposurerisk, and extended the regulatory mantle of Bank Negara beyond com-mercial banks to include finance companies and merchant banks. Creditrating agencies were established, and measures were taken to deepenfinancial markets – principally the inter-bank money market, the foreignexchange market, and in particular the stock market – with the establishmentof a Securities Commission in 1993, which led to improvements in tradingand settlement systems. The only financial market which remainedrelatively underdeveloped was the bond market, because of the surprisingretention of high minimum liquid asset requirements for banks (similar tothe Indian statutory liquidity ratio).

Rate deregulation accompanied prudential control. By February 1991,banking institutions were free to set deposit rates and a base lending rate


(BLR), subject to a standardized formula benchmarked on deposit rates.The lack of incentive in such a system for banks to improve their internalefficiency was corrected in November 1995, when banks became free toset a BLR subject to a ceiling, benchmarked on the three-month inter-bankrate. A cap of 4 per cent above BLR remained on the maximum lendingrate.

The sound configuration of capital account convertibility in Malaysiais especially important to keep in mind against the background of the capitalcontrols introduced after the external account crisis. These controls wereintroduced in a country that had a background of prudently liberal, ratherthan reckless, capital account openness.

The strength of Malaysian macro fundamentals has already beencommented on in section I. Malaysia was a country with high growth rates,of not less than 9 per cent in any year after 1988; the crisis of 1997 was thefirst year in a whole decade which saw growth dipping well below 9 percent, to 7.3 per cent. This growth record was accompanied by remarkablylow rates of inflation, at well under 5 per cent, a high savings rate in excessof 40 per cent, fiscal surpluses on current account, and remarkably lowlevels of external debt – both total and short-term – as a percentage ofexports.

C. The crisis of 1997

Table 12 shows the stability in the value of the ringgit against the USdollar in the 1990s. There were some upward pressures on the ringgit mid-decade, with a rise in its value from 2.7 ringgit to the dollar in 1990 to asteady 2.5 ringgit to the dollar in 1994–1996 – three years immediatelypreceding the crisis – despite high current account/GDP ratios.

For a country as open as Malaysia, with exports equal to or greaterthan GDP, the exchange rate is the most important price. For that price tobe plunged by forces beyond its control was a critical vulnerability thatthe country could not afford to ignore. Yet another vulnerability is the


ble

12

MA

LAYS

IA: E

XCH

AN

GE

RAT

ES A

ND

EXP

OR

TS, 1

990–

2000

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

Exch

ange

rate

(RM

/$)

End

per

iod

2.7

2.7

2.6

2.7

2.6

2.5

2.5

3.9

3.8

3.8

3.8

Per

iod

aver

age

2.7

2.8

2.6

2.6

2.6

2.5

2.5

2.8

3.9

3.8

3.8

Shar

es in

tota

l man

ufac

ture

dex

port

s (p

er c

ent)

Ele

ctro

nic

and

elec

trica

l tot

al56

.658

.058

.261

.563

.565

.765

.866

.568

.171

.872

.0(o

f whi

ch e

lect

roni

c):

32.8

31.8

32.5

34.4

35.4

38.6

40.8

45.1

48.0

53.3

53.0

Sem

icon

duct

ors

24.9

21.3

20.1

20.9

20.7

22.5

22.4

22.8

22.9

24.1

21.8

Ele

ctro

nic

equi

pmen

t7.

810

.512

.413

.514

.816

.018

.422

.325

.129

.231

.2To

tal m

anfd

. exp

orts

100.

010

0.0

100.

010

0.0

100.

010

0.0

100.

010

0.0

100.

010

0.0

100.

0

Sour

ce:

Ibid

., ta

bles

V.6

and

VIII

.5.

Not

e:Th

e fig

ures

for t

he y

ear 2

000

rela

te to

the

perio

d Ja

nuar

y–M

ay 2

000.

RM

=M

alay

sian

ring

git.


excessive dependence on electronics in manufactured exports, whichaccount for 85 per cent of the total export basket. Electronic and electricalexports accounted for more than half of total manufactured exports at thebeginning of the 1990s, and by end-decade the share had risen to morethan 70 per cent. Malaysia was not unique among its neighbours in thisrespect. The figures for the first half of 2000 show shares for electronicsexports of 58 per cent (Malaysia), 56 per cent (Singapore, excluding re-exports), 45 per cent (Taiwan Province of China) and 37 per cent (Republicof Korea).

A nominal depreciation of the ringgit to some degree was clearlynecessary to maintain export competitiveness at a time of steep currencydepreciation in the Asian competitive neighbourhood; but it was theaccompanying rise in domestic interest rates with the build-up of speculativedemand for the ringgit that was the major real threat, and clearly called forcontainment. The challenge posed was the need to reduce interest rateswithout precipitating a further decline in ringgit value.

After years of positive portfolio inflows, there was a net outflow in1997 of $10 billion (table 2), although FDI continued to hold steady, withinflows of $2.4 billion in 1997. Speculative pressure on the ringgit builtup, financed by ringgit obtained offshore. This was initially sought to becontained in August 1997 through an offer-side limit on currency swaps(sale of ringgit) with non-resident banks, fairly generously set at $2 millionper bank. This was a very limited response to the speculative pressures of1997, and the swap limit did very little to contain the slide in the ringgit(which fell to a low of 4.88/$ in January 1998) or the decline in foreignexchange reserves to $15 billion by end-December 1997.

Although contagion without any aggravation from weak fundamentalscould entirely have led to external crisis, there was an export decelerationin 1997, after sustained export growth since 1990 of about 18 per centannually in US dollars (figures in table 10). There also appears to havebeen some structural deterioration in the quality of bank loan portfolios inMalaysia as a result of rapid credit expansion and the collapse of the assetbubble, notwithstanding the controls in the system on loan quality. Table 13shows that after a long period of monotonic decline in the percentage of


ble

13

MA

LAYS

IA: F

INA

NC

IAL

SEC

TOR

IND

ICAT

OR

S, 1

990–

2000

(Per

cen

t)

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

Adj.

Adj.

Adj.

Adj.

Gro

ssN

etgr

oss

Net

gros

sN

etgr

oss

Net

gros

s

Non

-per

form

ing/

tota

l loa

ns (e

nd-p

erio

d)C

omm

erci

al b

anks

20.1

15.7

14.7

12.6

6.9

4.9

3.6

3.2

10.5

5.9

15.3

5.7

16.2

5.5

14.4

Fina

nce

com

pani

es21

.315

.815

.613

.09.

96.

64.

76.

513

.911

.928

.88.

623

.98.

923

.2M

erch

ant b

anks

12.6

8.7

6.9

5.0

9.5

7.8

1.7

3.5

10.6

10.9

22.5

12.3

27.0

12.5

27.8

Bank

ing

syst

em (o

vera

ll)20

.015

.414

.512

.37.

85.

53.

74.

19.

77.

518

.96.

618

.46.

516

.9

Com

mer

cial

ban

k in

tere

stra

tes

(ave

rage

yea

r-en

d)D

epos

it ra

te (1

yr.)

7.2

8.2

7.8

6.3

6.2

6.9

7.3

.9.

3.

5.7

.4.

0.

3.9

LIBO

R (1

yr.)

8.5

6.3

4.2

3.6

5.6

6.2

5.8

.6.

1.

5.5

.5.

7.

6.9

(Dep

. rat

e - L

IBO

R)

-1.2

1.9

3.6

2.7

0.6

0.6

1.5

.3.

3.

0.2

.-1

.7.

-3.0

Infla

tion

rate

3.1

4.4

4.7

3.6

3.7

3.4

3.5

.2.

7.

5.3

.2.

8.

1.3

(Rea

l dep

osit

rate

)4.

13.

83.

12.

72.

53.

53.

8.

6.6

.0.

4.

1.2

.2.

6

Lend

ing

rate

sBa

se (B

LR)

7.5

8.7

9.3

8.2

6.8

8.0

9.2

.10

.3.

8.0

.6.

8.

6.8

Aver

age

9.0

9.7

10.3

9.7

8.2

9.3

10.1

.11

.5.

9.7

.7.

8.

7.7

Sour

ce:

Ibid

., ta

bles

III.1

5-III

.18,

V.1

and

VI.1

3 fo

r Mal

aysi

a; IM

F, In

tern

atio

nal F

inan

cial

Sta

tistic

s fo

r LIB

OR

; ave

rage

Jan

uary

–May

for 2

000.

Not

e:•

The

NP

L fig

ures

for 2

000

rela

te to

end

-May

. The

inte

rest

rate

s fo

r 200

0 ar

e fo

r Jun

e 20

00.

•B

egin

ning

Dec

embe

r 199

7, n

on-p

erfo

rmin

g lo

ans

wer

e re

porte

d ne

t, as

a p

erce

nt o

f net

loan

s. T

o re

gain

com

para

bilit

y w

ith e

arlie

r yea

rs, t

he a

djus

ted

gros

ssh

are

of g

ross

loan

s is

con

stru

cted

from

repo

rted

figur

es o

f pro

visi

onin

g. F

urth

er, i

n th

e fir

st s

ix m

onth

s of

199

8, th

e pe

riod

of n

on-p

erfo

rman

ce fo

r the

repo

rted

figur

es w

as s

horte

ned

from

six

to th

ree

mon

ths.

Afte

r mid

-199

8, b

oth

sets

of f

igur

es a

re re

porte

d. T

he fi

gure

s in

this

tabl

e fo

r 199

8 an

d su

bseq

uent

year

s ar

e th

e si

x-m

onth

figu

res,

to re

tain

com

para

bilit

y w

ith e

arlie

r yea

rs.

•Th

e in

flatio

n ra

te is

from

the

Con

sum

er P

rice

Inde

x (C

PI).


non-performing loans (NPLs) in the banking system as a whole, from 20 percent in 1990 to 3.7 per cent in 1996, there was a rise to 4.1 per cent in thereported figure for 1997. This figure, however, understates the true increasebecause of the switch in the official reporting system from gross to netshares starting in December 1997 (NPLs also saw other definitionalvolatility over the period; see notes to table 13). Using the officially reportedfigures for loan provisioning, the adjusted gross share at end-1997 wasclose to 10 per cent, and by end-1998 this had risen to 19 per cent (see alsoGhani and Sood, 1999).

The early effects of depreciation pressures and ringgit selling werefelt in the offshore ringgit market. Speculative demand for ringgit forconversion into dollars in anticipation of a crash in ringgit value led to adramatic rise in offshore ringgit deposit rates, which by April 1998 hadrisen above 30 per cent. This was transmitted to the domestic market throughcapital outflows in response to the rise in offshore ringgit rates.

Domestic ringgit interest rates rose in response (table 13). The 12-monthdeposit rate towards end-1997, at 9.3 per cent, offered a real return todepositors of 6.6 per cent over what remained a low inflation rate, and theexcess over the 12-month dollar LIBOR rose to 3.3 per cent. The averagelending rate by the end of 1997 had risen to 11.5 per cent, in a countrywhere the average lending rate had not risen above 10 per cent after financialrestructuring, and this despite the November 1995 freeing of the BLR ofbanks subject to a cap (see appendix B). Real contraction followed. Thereal growth rate of the economy in 1998 worsened steadily to -10.1 percent in the third quarter. This in turn led to a deterioration in bank loanportfolios. However, the differential impact of recession on the differentsubsectors of the banking system show that the incidence of structuralweakness within the system was not uniform. The increase in NPLs wassharper among merchant banks and finance companies than amongcommercial banks. By the end of 1998, finance companies and merchantbanks had (net reported) NPLs of nearly 12 and 11 per cent respectively,corresponding in gross terms to 29 and 22.5 per cent respectively (table 13,for the extended period up to 2000, see table 14).

INDIRA RAJARAMAN152

D. The capital controls of 1998–1999

The measures introduced on 1 September 1998 (see appendix B) weredesigned to:

• Kill the offshore ringgit market, by prohibiting transfer of funds intothe country from externally held ringgit accounts, except for invest-ment in Malaysia (excluding credit to residents) or purchase of goodsin Malaysia. Since the offshore ringgit market, like any offshoremarket, could only function through externally held ringgit accountsin correspondent banks in the territory of the currency, the measureimmediately rendered offshore ringgit deposits dysfunctional. Offshore

Table 14

MALAYSIA: SHARE OF PROPERTY SECTORIN NON-PERFORMING LOANS, 1998–2000

(Per cent)

December December March1998 1999 2000

Commercial banksConstruction 14.0 14.7 14.8Purchase of residential property 8.1 10.4 10.6Purchase of non-resid. property 6.4 6.9 7.1Real estate 7.0 5.9 6.3Total property 35.5 37.9 38.8

Finance companiesConstruction 8.6 13.6 16.8Purchase of residential property 5.5 8.3 9.2Purchase of non-resid. property 7.2 8.4 9.3Real estate 9.5 3.8 4.1Total property 30.9 34.0 39.4

Merchant banksConstruction 31.7 22.6 26.4Purchase of residential property 0.0 0.7 0.8Purchase of non-resid. property 1.3 2.0 2.6Real estate 8.5 11.1 11.3Total property 41.6 36.4 41.0

Source: Ibid., tables III.19-III.21.


banks required access to freely usable onshore ringgit bank accountsto match their ringgit liabilities, and the new ruling eliminated freeusability. Holders of these deposits were given a month within whichto repatriate them to Malaysia. But it is important to note that thiswas not a requirement so much as a consequence of the controls onexternal ringgit accounts. This eliminated a major source of ringgitfor speculative buying of US dollars in anticipation of a ringgit crash.A corollary measure, demonetization of large-denomination ringgitnotes was announced, and later followed through, so as to curb thecirculation of ringgit outside Malaysia.

• Close off access of non-residents to domestic ringgit sources byprohibiting ringgit credit facilities to them, requiring that all tradetransactions be settled in foreign currencies, and mandating authorizeddepository institutions through which alone transactions in ringgit-denominated financial assets became permissible.

• Close the offshore market in Malaysian shares conducted through amechanism known as the Central Limit Order Book (CLOB).

• Obstruct speculative outward capital flows by imposing the require-ment of prior approval for residents to invest abroad in any form, andsetting tight limits on exports of foreign currency by residents forother than valid current account purposes.

• Protect the value of the ringgit and shore up foreign exchange reservesby requiring repatriation of export proceeds within six months of thedate of export.

• Drive a further wedge between the foreign exchange market and theimperative need of the hour for monetary easing, and thus regainmonetary independence by imposing a 12-month stay on outflows ofexternal portfolio capital (only the principal, interest and dividendpayments were freely repatriable).

International rating agencies responded uniformly to the capital controlmeasures by downgrading Malaysia’s credit and sovereign risk ratings. As

INDIRA RAJARAMAN154

the success of the package became undeniable, it was attributed toenvironmental factors, among them:

• Adequacy of foreign exchange reserves; and

• The underlying strong fundamentals of the Malaysian economy.

What is notable about the package was its thoroughness, and the cleverchoice of policies designed to control foreign exchange outflows andspeculation against the ringgit by non-residents and residents. It was thusthe absence of discrimination against foreign investors, the clear evidenceof sound economic logic underlying the package, and the effectiveness ofits enforcement by Bank Negara, which accounted for the success of thecapital controls as much as the enabling environment and the underlyingstrength of the Malaysian economy.

The severing of the link between domestic interest rates and theexternal sector was a major motivation underlying the entire exercise. Thestatutory reserve ratio (SRR) was brought down sharply from 13.5 percent – to which it had been raised in 1996/1997 to contain liquidity in thesystem as part of an initially orthodox response to downward pressures onthe currency – to 4 per cent in 1998, where it now remains (table 15). Thebenchmark for setting the ceiling on the base lending rate (BLR) of banks,hitherto the three-month inter-bank rate,3 was further shifted to the BNMintervention rate, with the same formula as before, so as to enhance BNMleverage over lending rates, with the permissible margin above thebenchmark reduced from 2.5 to 2.25 percentage points. There was also areduction in the cap on the maximum lending rate, which was reduced forthe first time since financial deregulation, from a spread of 4 per cent abovethe BLR to 2.5 per cent. The impact of these reductions is clearly visiblein table 13. The average lending rate fell from 11.5 per cent at end-1997 to9.7 per cent by end-1998, and the real one-year deposit rate fell from 6.6 percent at end-1997 to 0.4 per cent by end-1998. With a further fall in inflationrates in subsequent years from what for Malaysia was a high inflation rateof 5.3 per cent in 1998, the real interest rate has subsequently risen to2.6 per cent today.


ble

15

MA

LAYS

IA: M

ON

ETA

RY C

ON

TRO

L LE

VER

S, 1

990–

2000

(Per

cen

t at y

ear e

nd)

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

Stat

utor

y re

serv

e ra

tio (S

RR

)6.

57.

58.

58.

511

.511

.513

.513

.54.

04.

04.

0

Liqu

idity

rat

ioC

omm

erci

al b

anks

requ

ired

17.0

17.0

17.0

17.0

17.0

17.0

17.0

17.0

15.0

15.0

15.0

actu

al18

.418

.018

.018

.017

.917

.520

.317

.817

.919

.619

.5Fi

nanc

e co

mpa

nies

requ

ired

10.0

10.0

10.0

10.0

10.0

10.0

10.0

10.0

10.0

10.0

10.0

actu

al12

.112

.012

.112

.012

.412

.513

.110

.114

.018

.716

.4M

erch

ant b

anks

requ

ired

10.0

10.0

10.0

10.0

10.0

10.0

10.0

10.0

10.0

10.0

10.0

actu

al16

.516

.817

.918

.315

.815

.318

.910

.020

.320

.224

.8

New

liqu

idity

fram

ewor

kC

omm

erci

al b

anks

< 1

wee

k..

....

....

....

..5.

03.

03.

01

wee

k–1

mon

th..

....

....

....

..7.

05.

05.

0Fi

nanc

e co

mpa

nies

< 1

wee

k..

....

....

....

..5.

05.

05.

01

wee

k–1

mon

th..

....

....

....

..15

.07.

07.

0M

erch

ant b

anks

< 1

wee

k..

....

....

....

..5.

03.

03.

01

wee

k–1

mon

th..

....

....

....

..15

.05.

05.

0

Sour

ce:

Ibid

., ta

bles

III.2

6–III

.28.

Not

e:Th

e 20

00 fi

gure

s pe

rtain

to e

nd-J

une.

The

SR

R is

a s

tatu

tory

cas

h re

serv

e to

be

mai

ntai

ned

with

Ban

k N

egar

a. U

nder

the

(old

) liq

uidi

ty ra

tio, t

he re

quire

men

tw

as a

n ov

eral

l per

cent

age

of e

ligib

le li

abili

ties.

By

Apr

il 20

00, 2

6 ba

nkin

g in

stitu

tions

had

cro

ssed

ove

r to

the

New

Liq

uidi

ty F

ram

ewor

k. F

igur

es o

f act

uals

unde

r the

old

liqu

idity

ratio

per

tain

onl

y to

thos

e ba

nkin

g in

stitu

tions

that

had

not

cro

ssed

ove

r.

INDIRA RAJARAMAN156

The reaction of received orthodoxy to the Malaysian package wasthat the Malaysian recovery was no stronger than that of other East Asiancountries. However, the incremental impact of the package can only beassessed with respect to a Malaysian counterfactual, since the othercountries differed in terms of their fundamentals at the time of the crisis. Aformal counterfactual exercise, which requires a quarterly computationalgeneral equilibrium (CGE) model for the Malaysian economy, is beyondthe scope of this paper. What is undeniable is the speed of real correctionin the economy, from -10.6 per cent in the last six months of 1998 to -1.5 percent in the first quarter of 1999, and positive growth in all subsequentquarters. Even critics of the Malaysian capital controls had to concedethat the reduction in interest rates helped contain the increase in NPLs ofthe banking system that would surely have been a feature of anycounterfactual scenario. Standard and Poor’s is reported to have estimatedthat the NPLs would have risen to 30 per cent if interest rates had notfallen as sharply as they did (IMF, 1999). Also, the Federation of MalaysianManufacturers is quoted as having reported that the exchange rate peg andreduced interest rates lowered corporate uncertainty and made businessplanning easier (IMF, 1999).

The measures of September 1998 imposed, inter alia, a 12-monthwaiting period for repatriation of proceeds from liquidation of externalportfolio investments. In order to pre-empt possible large-scale outflowsat the conclusion of that period in September 1999, the waiting period wasreplaced as of 15 February 1999 by a system of graduated exit levies, withseparate regimes for capital already in the country and for capital broughtin after that date. For capital already in the country, there was a price onexit inversely proportional to duration of stay within the earlier stipulatedperiod of 12 months. Therefore capital that had entered the country morethan a year previously, before 15 February 1998, was free to leave at azero exit price. For capital yet to come in, there was a levy only on profits,defined to exclude dividends and interest, also graduated by length of stay.In effect, profits for the purpose of the new Malaysian rules were definedto equal realized capital gains. Accompanying clarifications were issuedin respect of repatriation of funds relating to investment in immovableproperty and FDI. Also investments in the newly-established, over-the-counter share market, MESDAQ, were exempted.


Although the replacement of the earlier controls with exit levies wasinternationally welcomed, there was also criticism of the new package. Asa levy applicable only at the time of conversion of ringgit proceeds intoforeign exchange, and therefore not a capital gains tax, it could not beoffset through double taxation agreements. The 10 per cent levy on profits,even on funds invested for a period over 12 months, was seen as generallydiscouraging portfolio capital inflows, and equity investments in particular,since interest and dividends are exempted. The higher levy of 30 per cent,applicable on gains on investments of less than a year’s duration, attractedespecially heavy criticism on the grounds that potential investors wouldapply the higher levy rate of 30 per cent to all investments, regardless oftheir expected maturity, because of the “last in, first out” rule (IMF, 1999).On 21 September 1999, the higher levy was eliminated, leaving in placeonly a single rate of 10 per cent on capital gains regardless of duration ofinvestment. In a further relaxation, the 10 per cent levy on capital gainswas retained, effective from 1 February 2001, only for capital in the countryfor a duration of less than one year. This too was done away with on 2 May2001.

The very criticism directed at the new package helped identify whatwas good about it and, more importantly, underlined why it could prove ofenduring worth in reducing volatility in capital flows. It is true that thelevy reduced the expected rate of return on equity to foreign investors, andthus raised the required pre-levy rate of return needed relative to othermarkets. This was an intended effort to reduce casual entry into Malaysia,and to ensure that capital would enter only when the fundamentals justifiedthe expectation of a higher pre-levy rate of return. By December 1999,international rating agencies had begun restoring the credit rating forMalaysia. The final benediction was bestowed when the Malaysian marketwas included as a component of the Morgan Stanley Capital InternationalIndices in May 2000.

The institutional origin of the capital control measures is unclear. BankNegara believes in cooperation with the Government rather than inde-pendence, a tradition said to have been laid down by the first MalaysianGovernor of Bank Negara, Tun Ismail Mohamed Ali. The facts on record,which show that, effective 1 September 1998, there was a change in the

INDIRA RAJARAMAN158

leadership of the team charged with implementing the new package ofselective exchange controls, suggest that the measures may not haveoriginated with the central bank.

The exchange control measures of September 1998 (other than theprohibition on portfolio repatriation) remain substantially in place, exceptfor some relaxation in respect of credit facilities in ringgit by residents tonon-residents, which are now permissible as long as the credit facility isnot used to purchase immovable property in Malaysia, up to a limit of200,000 ringgit, well below the earlier pre-1998 limit of 5 million ringgit.

E. Financial sector reform

In terms of the institutional requirements of the effort to insulate thecountry on a long-term basis from recurrences of external volatility, it wasclear that financial restructuring would have to come first, given that thiswas the foremost structural weakness in what was otherwise a soundlymanaged macroeconomic scenario at the time of the crisis. The weaknessof finance companies, and merchant banks in particular as evidenced bythe much sharper increase in their non-performing loans (NPLs), wasalready referred to in section III.C.

Bank Negara has exhibited exemplary awareness of the need forinstitutional strengthening in the financial sector. Listed below are theavenues along which initiatives have already begun.

(i) Bank restructuring

Begun in early 1998 and continuing under a four-pronged approach:

• Danaharta, an asset management company, was established for pur-chase of NPLs, exchanged against issue of zero-coupon, government-guaranteed bonds with five-year maturity and an additional rolloveroption of five years; yields are approximately the same as MalaysianGovernment security yields.


• Danamodal, a scheme for recapitalization of banks, was introduced.It was a clearly needed accompaniment to Danaharta purchases ofNPLs at a discount from banks. Danamodal itself was financedthrough the generalized issue of zero-coupon bonds, in the firstinstance to banking institutions themselves, with secondary trad-ability among corporate houses, special status as Class-1 liquefiableassets under the new liquidity framework, and zero risk weight forcapital adequacy purposes.

• Bank consolidation. Initiated in January 1998 for finance companiesalone, this was extended to all banking institutions in July 1999.By 1 September 2000, 54 financial institutions were consolidatedinto 10 new entities led by hand-picked anchor banks. There was,however, a negative stock market reaction to some of these mergerson news of an element of official coercion (Straits Times, 2000).

• Corporate Debt Restructuring Agency (CDRC). This was set up toprovide a platform for borrowers and creditors to work out debtrestructuring schemes.

(ii) Banks capacities for risk management

Begun as far back as 1995, banks are being nudged towards in-housecontrol mechanisms for management of the additional dimensions of marketrisk involved in cross-border transactions, and evidence of an appropriateinternal control structure is a requirement for banks to use financialderivatives. Since March 1998, banks have been required to conductmonthly stress simulation tests under a variety of scenarios and to reportto Bank Negara on a quarterly basis.

(iii) Prudential limits on sectoral exposure

Notwithstanding the effort to shore up banks’ internal capacities forrisk assessment, there was an unquestioned need for limits on exposure toproperty lending, given the central role this had played in structuralweakening of the financial sector (for an excellent account of the gen-esis of bank credit flows to the property sector, see UNCTAD, 1998). In

INDIRA RAJARAMAN160

October 1995 limits on maximum financing margins on purchase of high-end properties were introduced for the first time (see BNM, 1999), whilenot controlling lending for construction – the supply end. Predictably,margins on purchase had to be abolished in 1998 in a bid to clear the excesssupply of high-end properties. A 20 per cent limit on total outstandingloans to the broad property sector was imposed in April 1997, just beforethe exchange rate crisis (the limit excluded low-end residential properties,infrastructure and industrial projects). It was only in January 1999 thatlending for the construction of high-end residential properties includingresorts, hotels, golf courses and commercial properties, was prohibitedaltogether. This response to a serious oversupply problem, if enacted fiveyears earlier, might possibly have averted the exchange rate crisis of 1997altogether.

(iv) Limits on connected lending

Effective 1 April 1998:

• Single customer credit is limited at 25 per cent of capital funds,down from 30 per cent.

• Country exposure limits are set on a bank-specific basis.

• Credit facilities on large loans, defined as those exceeding 10 percent of total capital, are limited to 50 per cent in aggregate.

• There is prohibition on lending to directors, officers and employees,and to firms in which such persons have an interest of more than 5per cent.

Effective 1 April 1999:

Lending to large shareholders with shares of 20 per cent or more inthe bank making the loan is prohibited. This was the first time thatlending to owners was explicitly prohibited, although it is thoughtthat since sizeable stakeholders are likely also to be directors, the1998 prohibition went most of the way.


(v) Prudential limits on liquidity risk

Beginning January 1999, a new liquidity framework based on amaturity ladder approach was introduced in place of the earlier overallliquidity requirement, with banking institutions given the freedom to crossover to the new system. By April 2000, 26 institutions had opted for it. Inessence, the new liquidity framework breaks down liquid assets by degreeof liquidity.

(vi) Prudential limits on cross-border risks

Limits to banks’ exposure to foreign currency loans have been set at50 per cent of equity capital. These are prudential limits without penalties,but are in any case so generously set that they are in no danger of beingcrossed, and could perhaps be tightened.

(vii) Incorporating risks in loan classification, provisioning, capitaladequacy and disclosure

• Off-balance-sheet items have been incorporated in loan classi-fication and provisioning requirements since March 1998.

• The minimum risk-weighted capital requirement is still 8 per cent.Capital adequacy requirements are expected to move towards abank-specific configuration as a function of internal controls andsectoral exposure.

• Quarterly public disclosure requirements imposed in early 1998were relaxed in September 1998 along with the capital controlsthen introduced. A return to tighter disclosure norms was considerednecessary for restoration of discipline and market confidence.

INDIRA RAJARAMAN162

F. Conclusions

Malaysia had so free a capital account regime leading up to the 1997crisis was so free that there was even an offshore market in ringgit, perhapsthe only case of an offshore market in an emerging market currency. Theadvantage the offshore ringgit market gave foreign investors, in terms offine-tuning risk-management through access to the more diversifiedfinancial hedging products and instruments available offshore, paled incomparison with the exchange rate turbulence it facilitated in 1997.Eliminating the offshore market had to be an essential feature of any policypackage designed to drive a wedge between the foreign exchange marketand the imperative monetary policy need at the time of the crisis, whichwas to lower interest rates. The rise in ringgit interest rates was transmittedonshore from offshore speculative borrowing of ringgit to fund movementinto dollars in anticipation of a crash in the ringgit value, and had potentiallydevastating consequences for the domestic real economy and for what wasalready a structurally weak financial sector, overexposed to propertylending. The offshore ringgit market has now been wiped out, and thus theprincipal intent of the capital control measures introduced in September1998 has been fully achieved. The frozen Central Limit Order Book (CLOB)offshore shares, a residual problem after closure of the offshore ringgitmarket, is on the way to resolution through an offer menu that accom-modates shareholders at all ends of the liquidity spectrum. There remainno loose ends on the offshore front.

The exchange controls introduced in September 1998 remain largelyin place with the single exception of the 12-month holding period onrepatriation of portfolio principal, which after September 1999 was reducedto just a 10 per cent levy on capital gains. This was further confined,effective February 2001, to capital repatriated after a duration of less thana year in the country. On 2 May 2001, even this vestigial hurdle on financialexit from Malaysia was removed altogether.

The overwhelming need for financial sector consolidation is fullyrecognized and being addressed. Until that process is complete, Malaysiawould be ill-advised to open up its financial borders again. The exchangecontrols still in place obstruct free access to ringgit for non-residents as


part of the effort to prevent resurrection of an offshore ringgit market.Free movement from ringgit into dollars for residents is still possible, butthese dollars must be held in foreign exchange accounts in Malaysia. Thedifference from the point of view of control over the capital account, isthat the foreign exchange accounts are held by banking institutions withinthe regulatory ambit of Bank Negara. What is not permitted is export ofdollars outside the country. An officially-approved foreign currencyoffshore banking centre has been developed in Labuan. Outward portfolioflows, whether from corporates or resident individuals, require approval,which is rarely granted. But as in all approval-driven systems, the barrierscould be relaxed over time without a formal change of regime.

The overhang of NPLs is being handled through a set of interconnectedagencies. Stepping back from the complexity of the arrangements, what isbeing attempted is assistance to banks in loan rescheduling and manage-ment, offloading of bad loans at a discount to an asset management agency,and recapitalization of the worst-hit banks through bonds subscribed to bystronger banks in the system. Other planks in the financial restructuringprogramme included the mergers of what were 54 banks into 10 units by1 September 2000, which provoked negative stock market reactions onaccount of there having been some measure of official coercion.

The share of gross NPLs has come down somewhat since the worstpeak at end-1998 in the banking system as a whole and in all componentsof it except merchant banks. The property sector continues to account for40 per cent of NPLs. The controls introduced in the system in 1999 toprohibit lending for construction of high-end properties came five yearstoo late to avert the financial sector softening that was a contributory, ifnot the precipitating, factor in the 1997 crisis. Controls on connectedlending, now in place, again came five years too late.

With the continued dollar peg on the ringgit, the need for developinga domestic market for hedging instruments has taken a back seat, althoughit is clear that such markets will have to be developed over the long term.It has to be remembered that the offshore ringgit market developed inresponse to the need of non-residents for hedging instruments at a timewhen import and export settlements were denominated in ringgit, so that

INDIRA RAJARAMAN164

resident importers and exporters were relieved from the need to hedge.With import and export settlements now denominated in dollars, as part ofthe package of exchange control measures of September 1998, there willeventually be a need for hedging instruments once the peg is removed.Statements by the Malaysian Prime Minister on the occasion of the country’sNational Day on 31 August 2000 suggest the ringgit peg is here to stay,notwithstanding market estimates that the peg undervalues the ringgit, andthat higher import prices lead to lower investment than would be the caseunder a ringgit float (the 1998 drop of 20 percentage points in the investmentrate continued into 1999). The larger need remains for Malaysia to adoptan exchange rate policy that carries greater flexibility and adaptability tomonetary policy requirements. While this is recognized, there is littleevidence that the ringgit will go off the peg, or even be re-pegged, in theforeseeable future.

IV. CONCLUSIONS

India and Malaysia share a similarity of capital account regimes todaythat would have been unthinkable a decade earlier. The capital controlsintroduced in Malaysia in response to the external liquidity crisis of 1997,and the gradual capital account liberalization in India, also motivated byan external liquidity crisis in 1991, have brought about a remarkableconvergence between the capital account regimes in the two countries. Inboth countries non-resident capital inflows enjoy full freedom ofrepatriation. Outward capital flows for corporate residents are permissible,but controlled within prescribed limits. All other outflows of capital fromresidents are banned. There is a lesson in this that carries validity beyondthe specifics of the two countries studied.

The advantages of free cross-border flows of capital and of access toa global savings pool remain unquestionably valid. However, in emergingmarkets with institutional weaknesses in the financial sector, it is dam-aging to focus on the gains of free capital flows without the institutionalconsolidation that would prevent a recurrence of episodes of volatility.


The purpose of this paper is not to deliver a verdict on whether full capitalaccount convertibility should be an eventual world objective, issues in anycase already ably addressed in Cooper (1998) and others. Rather, the paperseeks to demonstrate that the transition to full capital convertibility neednot necessarily be monotonic at all times, and that retention of rights ofsovereign control over policy with respect to resident capital is essentialfor macroeconomic control over the real sector, especially in developingcountries (see also Schneider, 2000).

India in 1990 did not have current account convertibility; the rupeewas administered at a moving basket-peg; inward FDI was possible onlyon a case-by-case approval basis, subject to equity caps, curtailments onfreedom of functioning beyond those caps, and dividend repatriation linkedto export obligations; portfolio inflows were unknown; and outward FDIby corporates, although not unknown, was subject to an extended case-by-case approval basis. All other outward capital flows were banned, withcriminal penalties for contravention of the ban. An informal hawala channelfor capital outflows thrived, notwithstanding these penalties, with dollarsupplies from diverted private remittances.

Malaysia in 1990 had had current account convertibility for over twodecades; the ringgit had been a floating currency for 17 years, and therewas a flourishing offshore ringgit market in Singapore; inward FDI andportfolio capital inflows were unfettered, with approval needed only forflotation of issues on foreign bourses; and outward capital flows were freefor corporates, except those with large domestic borrowings. Outwardcapital flows from non-corporate residents were further freed in 1994 aspart of an attempt to contain an inward capital surge.

The overall policy framework for management of the capital account,the exchange rate and macroeconomic policy

The present Malaysian capital account regime is remarkably similarto the Indian, despite their very diverse historical points of origin. Outwardinvestment for corporate residents is permissible up to a limit, and beyondthat with approval, again similar to the newly liberalized Indian regime;

INDIRA RAJARAMAN166

although in all approval-driven systems, it is the manner in which theapproval is exercised that determines the substance of the regime.

Inward capital flows into Malaysia continue to be free and unfettered,as always, for both FDI and portfolio investment. Flotation of issues onforeign bourses has always required government approval. Thus thereremains full capital account convertibility for non-residents. Among themeasures introduced in September 1998 was the ban on any use ofexternally-held ringgit accounts in domestic banks other than towardspurchases of Malaysian financial or real assets. This was introduced tocurb the flourishing offshore ringgit market, which fuelled speculativemovements in exchange markets in 1997 and 1998. To that degree, thefreedom earlier enjoyed by non-residents to operate ringgit-denominatedaccounts outside Malaysia has been curtailed. But non-residents remainfree to hold such accounts in Malaysia, subject to the ban on direct creditto residents from such accounts.

The legal regime in India for capital account transactions has seen amajor alteration, effective as of 1 June 2000, with the introduction of FEMA,under which contraventions of the ban on capital outflows from residentsno longer invite criminal penalties. The Act also grants first-time automaticapproval for outward FDI by corporates through stock swap options, subjectto sectoral caps and with possible approval beyond those caps. All otheroutflows of capital from residents remain banned. Non-resident capitalinflows enjoy full freedom of repatriation, and effective from March 2000,face very few sector-specific entry barriers. Flotation of shares on foreignbourses remains subject to government approval.

Policy in India towards the capital account over the 1990s has beensteady in terms of both direction (liberalization of capital inflows) andpace (gradual). The slow pace of liberalization of inflows in the capitalaccount was undoubtedly an outcome, to some degree, of buoyant inflowsof current invisibles, private remittances in particular, in response to theearly moves towards current account liberalization. The routing of privateremittances through official channels, away from hawala, was an outcome,inter alia, of lifting the ban on gold imports. Other elements of currentaccount liberalization, by making possible legitimate purchases of foreign


currency, brought back flight capital through the mix of current and capitalaccount channels described in section II. However, no objective estimatesof the quantum of returning flight capital are available. With the return offlight capital, current account convertibility more than paid for itself.External reserves are now in the region of $40 billion (approximately8 months’ import cover), up from $1 billion, the crisis level in June 1991,immediately preceding reform.

Malaysian policy during the 1990s, from a starting point of free capitalaccount convertibility, has also demonstrated steadiness of purpose.Whether during the 1993 capital surge, or the 1997 capital outflows,Malaysian policy has asserted the sovereign right to impose temporaryreversals of freedom of cross-border capital flows, and thus to have capitalaccount convertibility on a reversible, rather than an irreversible, basis.During the 1993 capital inflow surge, which had a large short-termcomponent, restrictions were imposed on external borrowing, portfolioinvestment in Malaysian securities, and forward and swap transactions bybanks. These measures were reversed within a period of 8–12 months aspressures on the ringgit eased, with an outflow of short-term capital. ThusMalaysia rode out the inflow surge without the domestic monetaryturbulence of trying to sterilize free inflows.

The measures imposed in September 1998 have had a longer duration,and continue to remain in place, except for the phased reversal over a yearof the controls on portfolio repatriation. With the exception of the vestigial10 per cent levy on profits on portfolio capital in the country for less thanone year, which was removed in May 2001, and the successful killing ofthe offshore ringgit market through limits placed on permissible uses ofnon-resident accounts, these measures have curbed freedom of cross-borderflows for residents rather than non-residents. Although current accounttransactions do not carry category-specific caps, the tight limit imposed inSeptember 1998 on export of foreign currency by residents functions as auniform cap on outflows on both current and capital accounts. Malaysiahas decisively established its right to manage its capital account in such away as to impose a wedge if need arises, between the external value of theringgit and its sovereignty over domestic monetary policy.

INDIRA RAJARAMAN168

The exchange value of the ringgit remains at the peg of $1 = 3.8 ringgit,introduced in September 1998. The paradox is that the ringgit peg waspart of a package of measures designed to spur real growth through lowerdomestic interest rates – which had risen in response to speculative demandfor ringgit – and that today the peg undervalues the ringgit, so that pricesof imported capital goods are higher and investment lower than would bethe case in the absence of the peg. There is clearly a longer-term need toalign exchange rate policy with monetary policy. Even if this is recognizedwithin Bank Negara, the central bank is not free to introduce a unilateralchange of regime. At the same time, there is no expressed need for suchindependence. Macroeconomic policy is seen as an interlinked whole, tobe worked out in cooperation between the central bank and the executivearm of the Government.

The Indian rupee is market-determined in respect of trends. No otherpolicy is possible, given the inflation differential between India and itsmajor trading partners. However, the foreign exchange market remainsthin, and there are anywhere from one to three episodes of volatility in atypical year, in which the Reserve Bank of India (RBI) intervenes activelythrough alteration of liquidity conditions in the call-money market. Withthe introduction by the RBI of a Liquidity Adjustment Facility, and itscontinued efforts to improve depth in the call-money market, theseinterventions might become less jerky and ad hoc than they have been,noticeably during a major episode in mid-2000.

There is a coherent process of financial sector consolidation in placein India, and official recognition of the need for fiscal tightening, withoutwhich no lowering of domestic interest rates can take place. Financial sectorreforms remain snagged in the absence of any progress for establishing alegal framework for speedy liquidation of loss-making enterprises, whichin turn requires decisions that are difficult in the present political scenarioof coalition government.

Financial sector consolidation in Malaysia has received very carefulpolicy attention; it was begun in early 1998, well before the capital controlmeasures that came in September, with a four-pronged approach encom-passing asset management, bank recapitalization, loan management and


debt restructuring, and bank consolidation. There is a clear recognition ofthe damage done by uncontrolled lending to the property sector, whichwas finally halted in 1999 at the supply end after some unfruitful fumblingwith demand-end controls. Controls were also introduced in 1998 and 1999on connected lending. Impressive attempts are under way to improve banks’in-house capacities for management of credit and market risk, with banksrequired to conduct monthly stress simulation tests under a variety ofscenarios, and to report to Bank Negara at quarterly intervals. All of thesewould be promising were it not for occasional evidence of coercion inpolicy implementation, as, for example, in the ongoing process of bankconsolidation – where the mergers into 10 consolidated entities may havebrought together unwilling partners – and the possibility that enforcedsubscription to bonds for recapitalization of the worst-hit banks might carryadverse incentives for well-managed banks in the system (for other issuesthat arise in connection with banking regulation in relation to a newfinancial architecture, see Cornford, 1999).

The Extent to which the different regimes in recent years have contributedto more stable capital movements

Short-term debt as a percentage of the total debt stock has shown amarked fall over the 1990s in India, particularly after the 1997 changesdiscouraging short maturities in the deposit schemes offered to non-residentIndians. These deposit schemes, a peculiar feature of the Indian capitalaccount, have now moved to a more sensible configuration of deposit rates,benchmarked to LIBOR for foreign currency deposits, in place of theprevious configuration which offered arbitrage opportunities bordering onthe scandalous. With the new maturity floors on non-resident Indian depositsand non-trade commercial debt, short-term debt will henceforth be purelytrade-related.

To the extent that Indian liberalization of capital inflows has so farmoved at a faster pace for portfolio investments than for FDI, the regimehas not contributed to stable capital movements as much as if FDI hadbeen more rapidly facilitated. Portfolio capital is inherently volatile,although this has been retarded to some degree in India by the proceduraldifficulties of entry and exit. However, with dematerialized trading finally

INDIRA RAJARAMAN170

having captured the bulk of traded stocks, there has been a dramatic fall intransaction costs of both entry and exit. While this is good in general forreduction of friction in the system, it might lead to greater volatility inportfolio flows, in much the same way in which the liquidity of Malaysiancapital markets, both onshore and offshore, facilitated contagion. Fortu-nately, there was a further liberalization of entry for FDI in March 2000,with the move to a small negative list of sectors needing approval, insteadof, as previously, a listing of open sectors.

Malaysian short-term debt as a share of total debt is higher than forIndia, but normalized to exports it has historically been far lower. However,with the decline in Indian short-term debt, the two are now roughly at par.Malaysia unquestionably stabilized capital inflows during the mid-decadesurge, with entry obstructions that reversed the heavy short-term inflowsin particular, and with liberalization of portfolio outflows for residents.The regime change in 1998, however, was not directed principally atexternal capital, since it was introduced in September 1998 after portfoliooutflows of $10 billion had already taken place in 1997. At the time, it wasdesigned principally to stabilize the ringgit and, by imposing barriers toexit of resident capital, to sever the link between exchange markets andthe monetary policy need of the hour, which was to lower interest rates. Tothe extent that lower interest rates succeeded in containing the recessionof 1998 and facilitating real growth of 5.6 per cent in 1999, the measuressucceeded in stabilizing the macroeconomic system, and, as a corollary,sustained the stability of FDI inflows within the neighbourhood of $2 billionthroughout 1998 and 1999.

Thus the Malaysian regime in place since 1998 has achieved greaterstabilization of capital flows than what would have obtained in thecounterfactual, but a historically different configuration of policies mightpossibly have secured greater insulation against contagion in the first place.Thus the development of a market in foreign exchange derivatives inMalaysia would have pre-empted the development of the offshore ringgitmarket, which arose in response to the hedging needs of counterparties toringgit-denominated trade with Malaysia; the offshore market was wherenon-residents were able to obtain ringgit-denominated credit for speculationduring the currency crisis outside the reach of controls applicable on credit


to non-residents within Malaysia. Again, the controls on lending forconstruction of upmarket resorts and commercial properties, if enacted in1994 rather than in 1999, would have prevented the financial sectorsoftening that, along with export deceleration, precipitated capital outflowsin 1997. But after the onset of the crisis, the Malaysian response was awell-judged, macro-stabilization package with a focus on the real sectorcorrective called for at the time.

Likely future direction of policy and major influences

Malaysia is politically stable, with a Prime Minister firmly in controlof economic policy, and not due to face parliamentary elections until 2004.It is likely that the capital controls will remain in place until then. On theoccasion of Malaysia’s National Day on 31 August 2000, the Prime Ministerreaffirmed his commitment to the peg, despite widespread consensus thatthe ringgit is undervalued at the peg. There is worry that FDI may fall offin some years, and that the continuation of inflows at around the $2 billionlevel seen right through 1998 and 1999 merely represented a follow-throughon projects begun before the crisis. Thus, although the medium-term outlookis stable, both politically and in terms of the capital account regime, thereis an underlying apprehension about the peg and the eventual need for aregime that permits mutual compatibility between exchange rate andmonetary policy.

The likely economic policy direction in India is somewhat less certainthan in Malaysia because of the greater political uncertainty, despiteagreement across the political spectrum on the obstacles that must beaddressed if economic fundamentals are to be strengthened: the fiscaloverhang and the need for investment in infrastructure. There is broad-spectrum consensus that external capital must be wooed for infrastructureand other investments, so that the opening up to capital inflows of the1990s will continue regardless of the political configuration. What remainimpaled on the political scenario, however, are the ideologically-sensitivedecisions on cutting fiscal subsidies, which are critically necessary forcutting interest rates and thus for facilitating the real growth without whichno sustainable financial sector reform is possible. The import liberali-zation, which will be fully completed according to the WTO timetable by

INDIRA RAJARAMAN172

March 2001, has left in its wake defeated domestic competitors, and thelegal reform necessary to enable liquidation of loss-making enterprises isa function of the political strength of pro-reform elements in theGovernment. The national Government, if it lives out its full term, is duefor a change only in 2004. Cutting fiscal subsidies on petroleum products,food and electricity will raise cost-push inflation even as it engenders realgrowth (through reduction in the fiscal deficit and thereby domestic interestrates), so that with or without fiscal reform there is no immediate prospectof achieving the macroeconomic markers necessary for full convertibilityon capital account.


APPENDIX A

EXTERNAL POLICY MILESTONES: INDIA

Exchange rate regime/current account

1 March 1992: Dual exchange rates, administered/market-determined,under the Liberalized Exchange Rate ManagementSystem (LERMS).

1 March 1993: Unification of dual exchange rates into single-market-determined rate.

20 Aug. 1994: Current account convertibility (IMF Article VIII), withnotified, category-specific caps on outflows.

9 Jan. 1997: Caps on trade-related outflows removed.

31 March 2000: Quantitative restrictions removed on 714 out of 1,429imported items; remainder slated for removal by31 March 2001.

Capital account

Institutional/legal framework

30 Jan. 1992: Statutory empowerment of Securities Exchange Boardof India (SEBI) for regulation of stock markets.

8 Jan. 1993: Major alterations in Foreign Exchange Regulation Act(FERA) of 1973 granting parity of status to foreign andIndian-owned companies, and liberalizing outwardinvestments by Indian companies in joint venturesoverseas.

INDIRA RAJARAMAN174

23 July 1996: Legislation passed in Parliament for setting up of theNational Stock Depository enabling first-time paperlesstrading on the capital market.

30 May 1997: Report of the (Tarapore) Committee on Capital AccountConvertibility (CAC) recommending a three-year phasedmove to CAC, subject to macro targets: Gross fiscaldeficit/GDP 3.5 per cent (1999–2000); inflation rates3–5 per cent (1997–2000); debt service ratio of 20 percent (1999–2000).

5 April 1999: All trading in India’s two main stock indices, Nifty(NSE–50 stock index) and Sensex (BSE-30 stock index)dematerialized.

1 June 2000: Replacement of FERA by Foreign Exchange Manage-ment Act (FEMA); contraventions hereafter to be dealtwith under civil, not criminal, law.

Foreign direct investment

24 July 1991: Under new industrial policy, first-time automatic approval(with export obligations) of FDI up to 51 per cent in34 specified sectors, higher than 51 per cent permissiblewith approval; in place of earlier case-by-case approvalsubject to 40 per cent ceiling in all but high-technologyor export-oriented projects.

13 April 1992: India signs Multilateral Investment Guarantee AgencyProtocol (MIGA) for protection of foreign investments.

31 March 2000: All FDI placed under automatic route except for a smallnegative list for sectors still requiring licensing (forboth domestic and foreign investment) or for foreigninvestment exceeding notified sectoral caps.


Foreign equity inflows

Foreign Institutional Investors (FIIs)

30 Jan. 1992: FIIs allowed to invest with full repatriability of principaland income in primary/secondary markets, subject toregistration with SEBI; aggregate ceiling of 24 per centof issued share capital; and individual ceiling of 5 percent.

22 July 1996: Individual ceiling raised from 5 to 10 per cent.

4 April 1997: Aggregate ceiling raised from 24 to 30 per cent.

24 April 2000: Aggregate ceiling raised from 30 to 40 per cent.

Capital issues on foreign bourses

1 April 1992: a Indian companies permitted to issue, subject to govern-ment approval, Foreign Currency Convertible Bonds andordinary shares through Global/American DepositoryReceipts (GDRs and ADRs) on Overseas Stock/Over theCounter Exchanges, with full repatriation benefits andno lock-in period, but with end-use restrictions.

22 May 1998: All end-use restrictions lifted except for ban on use ofGDR/ ADR issue proceeds for investment in real estate/stock market.

External commercial borrowing (ECB)

Subject to an overall annual ceiling with preference for infrastructure andexport sector financing, and restrictions on utilization for rupee expenditureb

relaxed for:

INDIRA RAJARAMAN176

18 May 1995: Manufacturing companies:Limit: $1 millionMinimum maturity: 3 years

8 Jan. 1996: Non-manufacturing companies:Same limits as for manufacturing

19 June 1996: Limit: $3 million

31 March 1997: Long-term limit : $100–$200 millionMinimum maturity : 10–20 years.

31 March 1997: Inward remittance of funds for imports permissible withutilization lag of up to one year.

16 June 1998: Loans with minimum average maturity of 10 years out-side aggregate cap on ECB.

22 May 1998: No end-use restrictions. Minimum maturity between3 years (simple) to 5 years (average) varying directlywith amount borrowed; 8–16 years for long-term win-dow.

Forex financial c markets

7 April 1997: Scrapping of CRR and SLR on inter-bank borrowingsleads to MIBOR.

15 April 1997: 1. First-time permission for forward foreign exchangecontracts without documentary evidence of underlyingexposure, and beyond six months; subject to a decla-ration of exposure supported by average export/importturnover of last two years.

2. Case-by-case approval of rupee/foreign currencyswaps replaced by permission for authorized dealersto operate “swap book” within their open positionlimits.


11 June 1998: First-time permission for forward exchange cover to FIIsto the extent of 15 per cent of outstanding investmentsas on that date.

24 April 1999: Limit for forward cover: 15 per cent of investments ason 31 March 1999 (with utilization, further extension ofcover possible); entire incremental investment thereafter.

Foreign currency holdings (banned in general for residents)

1 April 1993: Exchange Earners Foreign Currency (EEFC) Schemeunder which exporters permitted to retain part of theirearnings in accounts held abroad.

14 Aug. 2000: Halving of EEFC balances effective 23 August 2000.

10 Oct. 2000: EEFC ruling reversed.

Source: RBI, Annual Reports (various).a The notification was issued on 12 November 1993, but with backdated

effect from 1 April 1992.b As distinct from use for import financing; rupee expenditure on

investment in stock markets/real estate is prohibited.c Deregulation dates for domestic bank deposit rates are given in notes to

table 7, along with deregulation details on NRI deposits.

INDIRA RAJARAMAN178

APPENDIX B

EXTERNAL POLICY MILESTONES: MALAYSIA

Exchange rate regime/current account

11 Nov. 1968: Current account convertibility (IMF Article VIII).

8 May 1973: Currency (Malaysian ringgit) floated. Offshore ringgitmarket develops over time, mainly in Singapore, as aresult of imports and exports denominated in ringgit, tomeet hedging needs of non-resident counterparties.

2 Sep. 1998: Ringgit pegged at $1 = 3.8 ringgit

Capital account

1973; 1987: Capital inflows and outflows liberalized.

27 Feb. 1994: Relaxation of restrictions on outward portfolio investment.

2 Jan. 1990: Over-the-counter (CLOB) market in Malaysian sharesopens, giving further impetus to offshore ringgit market inSingapore.

1989–1995: Reforms of financial sector accompanying capital accountliberalization:

Banking and Financial Institutions Act (BAFIA), 1989,extending Bank Negara Malaysia (BNM) prudentialregulation to finance companies and merchant banks inaddition to commercial banks.

February 1991

1. Banking institutions free to set deposit rates.


2. Base lending rate (BLR) of each bank subject tostandardized formula benchmarked on deposit rates.

3. Maximum lending rate capped at 4 per cent above BLR.

November 1995

Banks free to set BLR subject to ceiling (benchmarkedon 3-month interbank rate + 2.5 per cent margin).

Inward capital controls (temporary)

17 Jan. 1994–20 Jan. 1995: Ceiling on foreign borrowing by domestic banks (ex-

cluding trade-related and direct investment).24 Jan.–12 Aug. 1994: Restrictions on portfolio investment in Malaysian secu-

rities.23 Feb.–16 Aug. 1994: Restrictions on forward and swap transactions by banks.

Controls on outward capital/closure of non-resident accessto ringgit

4 Aug. 1997: Currency swap limit of $2 million per bank group onoffer side (sale of ringgit) on transactions other thancurrent account with non-residents.

1 Sep. 1998: 1. Prohibition on transfer of funds from externally-heldringgit accounts except for:

i. Purchase of ringgit-denominated assets includingimmovable property in Malaysia but excludingringgit credit to residents.

ii. Expenditure on administrative expenses/purchaseof goods and services in Malaysia. With the conse-quent immobilization of offshore ringgit deposits,holders of offshore ringgit holdings were permitted

INDIRA RAJARAMAN180

to repatriate to Malaysia freely by 1 October; withpermission thereafter.

2. Prohibition on ringgit credit facilities by residents tonon-residents (previously subject to a limit).

3. All transactions in ringgit-denominated financialassets permissible only through authorized depositoryinstitutions.

4. All trade transactions to be settled in foreign currency.

5. Closure, with effect from 16 September 1998 of off-shore market in Malaysian equities (CLOB), thusfreezing holdings of 172,000 investors, worth 10 bil-lion ringgit, in 112 Malaysian corporations.

6. Prior approval beyond a certain limit for all residentsto invest abroad in any form (previously applicableonly to domestic corporates with domestic borrowing).

7. Limits on exports of foreign currency by residents.

8. All export proceeds to be repatriated to Malaysiawithin six months from date of export.

9. 12-month waiting period for conversion to dollars ofringgit proceeds from sale of Malaysian securities;no limits on repatriation of interest, dividends.

15 Feb. 1999: Ban on portfolio repatriation replaced by graduated exitlevies decreasing with duration of investment:

1. Investments made prior to 15 February 1999 (only onprincipal; no levy on dividends and interest):

Duration of investment Levy rate (percentage)

< 7 months 307–9 months 20

9–12 months 10> 12 months 0


2. Investments after 15 February 1999 (only on capitalgains; no levy on principal, dividends or interest):

Duration of investment Levy rate

< 12 months 30> 12 months 10

21 Sep.1999: Flat 10 per cent exit levy on capital gains on portfolioinvestment irrespective of duration of investment.

1 Feb. 2001: 10 per cent exit levy applicable only on duration ofinvestment < 12 months removed effective 2 May 2001.

Related changes not directly targeted at capital flows

12 Sep. 1998: Demonetization of large-denomination ringgit notes(1,000 ringgit and 500 ringgit) effective 1 July 1999.

1 Sep. 1998: 1. Maximum BLR benchmarked on BNM interventionrate in place of inter-bank rate, with 2.25 per centmargin.

2. Maximum margin above BLR of banks lowered from4 to 2.5 percentage points.

3. Enactment of Pengurusan Danaharta Nasional BerhadAct to set up Danaharta, an asset management company.

Source: BNM (1999) and BNM, Annual Reports (various) supplemented by IMF(1999).

INDIRA RAJARAMAN182

NOTES

1 A further one-off $1 billion inflow into private transfers in 1996/97 was from redemptionto residents in India of the India Development Bonds floated in the late 1980s.

2 The Resurgent India Bonds (RIB) issued by India in late 1998, which raised in excessof $4 billion, have not been entered in the BIS aggregate for debt securities.

3 The benchmark was actually 0.8 (three-month inter-bank rate)/(1-SRR) as reported byBNM (1999).

REFERENCES

Athukorala P (1998). Swimming against the tide: Crisis management in Malaysia. ASEANEconomic Bulletin, 15: 3 (December): 281–289.

Bank Negara Malaysia (1999). The Central Bank and the Financial System in Malaysia:A Decade of Change, 1989–1999. Kuala Lumpur, Bank Negara Malaysia.

Bank Negara Malaysia (2000). Monthly Statistical Bulletin, June. Kuala Lumpur, Bank Negara,Malaysia.

Bank Negara Malaysia (various) Annual Reports, 1997, 1998 and 1999. Kuala Lumpur, BankNegara, Malaysia.

Borgini P, Claessens S and Ferri G (2000). Political economy of distress in East Asian financialinstitutions. World Bank Policy Research Working Paper 2265. Washington, DC, WorldBank.

Brown SJ, Goetzmann WN and Park J (1998). Hedge funds and the Asian currency crisis of1997. Working paper no. 6427, National Bureau of Economic Research, Cambridge, MA.

Cole DS, Scott HS and Wellons PA (1995). Asian Money Markets. New York, Oxford UniversityPress.

Cooper R (1998). Should capital account convertibility be a world objective? In: Kenen P,ed., Should the IMF Pursue Capital Account Convertibility? Princeton Essays inInternational Finance. 207 (May): 11–19.

Cornford A (1999) Some remarks on banking regulation and supervision and their limitationsin relation to a new financial architecture. Paper presented at the Conference on Structure,Instability and the World Economy: Reflections on the Economics of Hyman P. Minsky.The Jerome Levy Economics Institute, Blithewood, NY, 21–23 April.

Das T (1999). East Asian economic crisis and lessons for debt management. In: Vasudevan A,ed., External Debt Management: Issues, Lessons and Preventive Measures. Mumbai,Reserve Bank of India: 77–95.

Furman J and Stiglitz J (1998). Economic crises: Evidence and insights from East Asia.Brookings Papers on Economic Activity. 2: 1–135.

Ghani E and Sood V (1999). Productivity growth, capital accumulation and the banking sector:Some lessons from Malaysia. World Bank Policy Research Working Paper, 2252.


Government of India (1991) Report of the (Narasimham) Committee on Financial System.Government of India (1998). Report of the (Narasimham) Committee on Banking Sector

Reforms. New Delhi.Government of India (2000). Economic Survey, 1999–2000. New Delhi.International Monetary Fund (various). International Financial Statistics. Washington, DC,

IMF.International Monetary Fund (1999). Malaysia: selected issues. IMF Staff Country Report,

99/86. Mimeo. Washington, DC, IMF, August.James H (1999). Is liberalization reversible? Finance and Development, 36: 4, December.Radelet S and Sachs JD (1998). The East Asian financial crisis: Diagnosis, remedies, prospects.

Brookings Papers on Economic Activity, 1: 1–74.Rajaraman I (1991). Trade-relevant external value of the rupee: 1974–89. Journal of Foreign

Exchange and International Finance, 4: 284–299, Oct.–Dec.Rajaraman I (1997). A profile of economic reform in India. Development Advisory Group

Paper 22, School of Public Policy, University of Birmingham, Birmingham, UnitedKingdom, July.

Rajaraman I, Bhaumik S and Bhatia N (1999). NPA variations across Indian commercial banks:Some findings. Economic and Political Weekly, XXXIV: 161–168, 3–4:January.

Reddy, YV (2000). Monetary and Financial Sector Reform in India. New Delhi, UBS Publisher.Reserve Bank of India (1997). Report of the (Tarapore) Committee on Capital Account

Convertibility . Mumbai.Reserve Bank of India (1999). Handbook of Statistics on Indian Economy.Reserve Bank of India (various). Annual Reports. Assorted issues.Reserve Bank of India (various). Reports on Currency and Finance. Assorted issues.Reserve Bank of India (various). Monthly Bulletin. Assorted issues.Reserve Bank of India. Report on Trends and Progress of Banking in India, 1999–2000.Schneider B (2000). Issues in Capital Account Convertibility in Developing Countries. Mimeo.

London, Overseas Development Institute, June.Straits Times (2000). KL stocks sink to their lowest this year. 2 September: 4.Summers LH (2000). International financial crises: Causes, prevention, and cures American

Economic Review, 90 (2): 1–16, May.UNCTAD (1998) Trade and Development Report 1998. New York and Geneva, United Nations,

United Nations publications sales no. E.98.II.D.6.Williamson J and Mahar M (1998). A review of financial liberalization. South Asia Region

Internal Discussion Paper, Washington, DC, World Bank, January.

new management of capital flowsunctad.org/en/docs/gdsmdpb20031p3_en.pdf · 2012. 6. 18. ·...

Documents